{
  "schemaVersion": 3,
  "updated": "2026-10-04",
  "selection": {
    "minimumRecentFamiliesPerLab": "2–3",
    "retainRecentMonths": 6,
    "refreshIntervalHours": 24,
    "policy": "Keep releases from the past six months plus each major lab’s latest two or three families. One result per family with size variants grouped. One GGUF publisher per source model; no duplicate conversions, base/instruct pairs, or community fine-tunes."
  },
  "models": [
    {
      "repo": "unsloth/Qwen3.8-Flash-Next-GGUF",
      "id": "qwen-qwen3-8-flash-next",
      "name": "Qwen3.8-Flash-Next",
      "sourceRepo": "Qwen/Qwen3.8-Flash-Next",
      "parameters": 176943899520,
      "layers": 48,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/de4b8e4d43b917e7706784d8bb445c9af86a3540/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 904004000
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 12,
            "bytesPerToken": 2048
          },
          {
            "layers": 12,
            "bytesPerToken": 256
          }
        ],
        "stateBytes": 119144448
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "other",
      "released": "2026-08-24",
      "downloads": 1441216,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "38bb39ee97821de2c9009abb7e93950eec396e66",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/de4b8e4d43b917e7706784d8bb445c9af86a3540/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 111334654784,
          "files": [
            "UD-Q4_K_XL/Qwen3.8-Flash-Next-UD-Q4_K_XL-00001-of-00004.gguf",
            "UD-Q4_K_XL/Qwen3.8-Flash-Next-UD-Q4_K_XL-00002-of-00004.gguf",
            "UD-Q4_K_XL/Qwen3.8-Flash-Next-UD-Q4_K_XL-00003-of-00004.gguf",
            "UD-Q4_K_XL/Qwen3.8-Flash-Next-UD-Q4_K_XL-00004-of-00004.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 188225033248,
          "files": [
            "Q8_0/Qwen3.8-Flash-Next-Q8_0-00001-of-00006.gguf",
            "Q8_0/Qwen3.8-Flash-Next-Q8_0-00002-of-00006.gguf",
            "Q8_0/Qwen3.8-Flash-Next-Q8_0-00003-of-00006.gguf",
            "Q8_0/Qwen3.8-Flash-Next-Q8_0-00004-of-00006.gguf",
            "Q8_0/Qwen3.8-Flash-Next-Q8_0-00005-of-00006.gguf",
            "Q8_0/Qwen3.8-Flash-Next-Q8_0-00006-of-00006.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-8",
      "familyName": "Qwen 3.8",
      "likes": 5846,
      "sourceDownloads": 1404413,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.33065418179837797,
        "source": "https://huggingface.co/Qwen/Qwen3.8-Flash-Next"
      },
      "edition": "qwen3.8-flash-next",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.8-2.4T-A95B-GGUF",
      "id": "qwen-qwen3-8-2-4t-a95b",
      "name": "Qwen3.8-2.4T-A95B",
      "sourceRepo": "Qwen/Qwen3.8-2.4T-A95B",
      "parameters": 2446182725504,
      "layers": 92,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B/raw/207bd685a7e3696cfaff12ded7c6a7ea0f88c996/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 23,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 601423872
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-08-08",
      "downloads": 8404,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "567d3e6ac26c5474b18311e619c04350fb9a5556",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B/raw/207bd685a7e3696cfaff12ded7c6a7ea0f88c996/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-IQ4_XS",
          "precision": 4,
          "bytes": 1310876469952,
          "files": [
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00001-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00002-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00003-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00004-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00005-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00006-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00007-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00008-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00009-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00010-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00011-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00012-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00013-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00014-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00015-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00016-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00017-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00018-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00019-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00020-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00021-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00022-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00023-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00024-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00025-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00026-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00027-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00028-of-00029.gguf",
            "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-00029-of-00029.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 2600249563264,
          "files": [
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00001-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00002-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00003-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00004-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00005-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00006-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00007-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00008-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00009-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00010-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00011-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00012-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00013-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00014-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00015-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00016-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00017-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00018-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00019-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00020-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00021-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00022-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00023-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00024-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00025-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00026-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00027-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00028-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00029-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00030-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00031-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00032-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00033-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00034-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00035-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00036-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00037-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00038-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00039-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00040-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00041-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00042-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00043-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00044-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00045-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00046-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00047-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00048-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00049-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00050-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00051-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00052-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00053-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00054-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00055-of-00056.gguf",
            "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-00056-of-00056.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-8",
      "familyName": "Qwen 3.8",
      "likes": 1279,
      "sourceDownloads": 35650,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.03883601948845691,
        "activeParameters": 95000000000,
        "source": "https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B"
      },
      "edition": "qwen3.8-2.4t-a95b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.8-27B-GGUF",
      "id": "qwen-qwen3-8-27b",
      "name": "Qwen3.8-27B",
      "sourceRepo": "Qwen/Qwen3.8-27B",
      "parameters": 27320697856,
      "layers": 64,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.8-27B/raw/1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 927607488
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 16,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 158859264
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-08-05",
      "downloads": 6314973,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "4ca720788d1e01f1bff70c033e0d0028fd02e502",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.8-27B/raw/1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 16464440224,
          "files": [
            "Qwen3.8-27B-UD-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 29047086048,
          "files": [
            "Qwen3.8-27B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-8",
      "familyName": "Qwen 3.8",
      "likes": 16829,
      "sourceDownloads": 6895117,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "qwen3.8-27b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.6-27B-GGUF",
      "id": "qwen-qwen3-6-27b",
      "name": "Qwen3.6-27B",
      "sourceRepo": "Qwen/Qwen3.6-27B",
      "parameters": 26895998464,
      "layers": 64,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.6-27B/raw/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 927607360
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 16,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 158859264
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-04-21",
      "downloads": 784287,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "82d411acf4a06cfb8d9b073a5211bf410bfc29bf",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.6-27B/raw/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 16817244384,
          "files": [
            "Qwen3.6-27B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 28595763424,
          "files": [
            "Qwen3.6-27B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-6",
      "familyName": "Qwen 3.6",
      "likes": 2317,
      "sourceDownloads": 2451011,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "qwen3.6-27b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.6-35B-A3B-GGUF",
      "id": "qwen-qwen3-6-35b-a3b",
      "name": "Qwen3.6-35B-A3B",
      "sourceRepo": "Qwen/Qwen3.6-35B-A3B",
      "parameters": 34660610688,
      "layers": 40,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B/raw/995ad96eacd98c81ed38be0c5b274b04031597b0/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 899283680
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 10,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 66846720
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-04-15",
      "downloads": 1291994,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "a483e9e6cbd595906af30beda3187c2663a1118c",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B/raw/995ad96eacd98c81ed38be0c5b274b04031597b0/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 22134528992,
          "files": [
            "Qwen3.6-35B-A3B-UD-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 36903140320,
          "files": [
            "Qwen3.6-35B-A3B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-6",
      "familyName": "Qwen 3.6",
      "likes": 2909,
      "sourceDownloads": 3311432,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.08655358173012927,
        "activeParameters": 3000000000,
        "source": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B"
      },
      "edition": "qwen3.6-35b-a3b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-0.8B-GGUF",
      "id": "qwen-qwen3-5-0-8b",
      "name": "Qwen3.5-0.8B",
      "sourceRepo": "Qwen/Qwen3.5-0.8B",
      "parameters": 752393024,
      "layers": 24,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-0.8B/raw/2fc06364715b967f1860aea9cf38778875588b17/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 204987232
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 20643840
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-28",
      "downloads": 204822,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "6ab461498e2023f6e3c1baea90a8f0fe38ab64d0",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-0.8B/raw/2fc06364715b967f1860aea9cf38778875588b17/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 532517120,
          "files": [
            "Qwen3.5-0.8B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 811843840,
          "files": [
            "Qwen3.5-0.8B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 741,
      "sourceDownloads": 2585330,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "qwen3.5-0.8b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-2B-GGUF",
      "id": "qwen-qwen3-5-2b",
      "name": "Qwen3.5-2B",
      "sourceRepo": "Qwen/Qwen3.5-2B",
      "parameters": 1881825088,
      "layers": 24,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-2B/raw/15852e8c16360a2fea060d615a32b45270f8a8fc/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 668227264
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 20643840
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-28",
      "downloads": 316761,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "f6d5376be1edb4d416d56da11e5397a961aca8ae",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-2B/raw/15852e8c16360a2fea060d615a32b45270f8a8fc/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 1280835840,
          "files": [
            "Qwen3.5-2B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 2012012800,
          "files": [
            "Qwen3.5-2B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 423,
      "sourceDownloads": 4883930,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "qwen3.5-2b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-4B-GGUF",
      "id": "qwen-qwen3-5-4b",
      "name": "Qwen3.5-4B",
      "sourceRepo": "Qwen/Qwen3.5-4B",
      "parameters": 4205751296,
      "layers": 32,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-4B/raw/851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 672423616
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 8,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 53477376
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-27",
      "downloads": 1047505,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "e87f176479d0855a907a41277aca2f8ee7a09523",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-4B/raw/851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 2740937888,
          "files": [
            "Qwen3.5-4B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 4482403488,
          "files": [
            "Qwen3.5-4B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 996,
      "sourceDownloads": 7791996,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "qwen3.5-4b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-9B-GGUF",
      "id": "qwen-qwen3-5-9b",
      "name": "Qwen3.5-9B",
      "sourceRepo": "Qwen/Qwen3.5-9B",
      "parameters": 8953803264,
      "layers": 32,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-9B/raw/c202236235762e1c871ad0ccb60c8ee5ba337b9a/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 918166080
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 8,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 53477376
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-27",
      "downloads": 1288271,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "3885219b6810b007914f3a7950a8d1b469d598a5",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-9B/raw/c202236235762e1c871ad0ccb60c8ee5ba337b9a/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 5680522464,
          "files": [
            "Qwen3.5-9B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 9527502048,
          "files": [
            "Qwen3.5-9B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 2091,
      "sourceDownloads": 9047776,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "qwen3.5-9b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-122B-A10B-GGUF",
      "id": "qwen-qwen3-5-122b-a10b",
      "name": "Qwen3.5-122B-A10B",
      "sourceRepo": "Qwen/Qwen3.5-122B-A10B",
      "parameters": 122111526912,
      "layers": 48,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-122B-A10B/raw/dc4d348443bc740c68e2d77492492c11606384d5/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 908724960
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 12,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 158072832
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-24",
      "downloads": 156038,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "51eab4d59d53f573fb9206cb3ce613f1d0aa392b",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-122B-A10B/raw/dc4d348443bc740c68e2d77492492c11606384d5/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 76536964608,
          "files": [
            "Q4_K_M/Qwen3.5-122B-A10B-Q4_K_M-00001-of-00003.gguf",
            "Q4_K_M/Qwen3.5-122B-A10B-Q4_K_M-00002-of-00003.gguf",
            "Q4_K_M/Qwen3.5-122B-A10B-Q4_K_M-00003-of-00003.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 129871935104,
          "files": [
            "Q8_0/Qwen3.5-122B-A10B-Q8_0-00001-of-00004.gguf",
            "Q8_0/Qwen3.5-122B-A10B-Q8_0-00002-of-00004.gguf",
            "Q8_0/Qwen3.5-122B-A10B-Q8_0-00003-of-00004.gguf",
            "Q8_0/Qwen3.5-122B-A10B-Q8_0-00004-of-00004.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 624,
      "sourceDownloads": 512579,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.08189235081145556,
        "activeParameters": 10000000000,
        "source": "https://huggingface.co/Qwen/Qwen3.5-122B-A10B"
      },
      "edition": "qwen3.5-122b-a10b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-27B-GGUF",
      "id": "qwen-qwen3-5-27b",
      "name": "Qwen3.5-27B",
      "sourceRepo": "Qwen/Qwen3.5-27B",
      "parameters": 26895998464,
      "layers": 64,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-27B/raw/fc05daec18b0a78c049392ed2e771dde82bdf654/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 927607040
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 16,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 158859264
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-24",
      "downloads": 207131,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "3221f178a6b842d04f1fb42f1c413534adcc0a6a",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-27B/raw/fc05daec18b0a78c049392ed2e771dde82bdf654/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 16740812704,
          "files": [
            "Qwen3.5-27B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 28595763104,
          "files": [
            "Qwen3.5-27B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 1057,
      "sourceDownloads": 1871002,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "qwen3.5-27b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-35B-A3B-GGUF",
      "id": "qwen-qwen3-5-35b-a3b",
      "name": "Qwen3.5-35B-A3B",
      "sourceRepo": "Qwen/Qwen3.5-35B-A3B",
      "parameters": 34660610688,
      "layers": 40,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/raw/59d61f3ce65a6d9863b86d2e96597125219dc754/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 899283648
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 10,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 66846720
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-24",
      "downloads": 507496,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "bc014a17be43adabd7066b7a86075ff935c6a4e2",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/raw/59d61f3ce65a6d9863b86d2e96597125219dc754/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 22016023168,
          "files": [
            "Qwen3.5-35B-A3B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 36903139968,
          "files": [
            "Qwen3.5-35B-A3B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 1519,
      "sourceDownloads": 1587887,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.08655358173012927,
        "activeParameters": 3000000000,
        "source": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B"
      },
      "edition": "qwen3.5-35b-a3b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Qwen3.5-397B-A17B-GGUF",
      "id": "qwen-qwen3-5-397b-a17b",
      "name": "Qwen3.5-397B-A17B",
      "sourceRepo": "Qwen/Qwen3.5-397B-A17B",
      "parameters": 396346350336,
      "layers": 60,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/Qwen/Qwen3.5-397B-A17B/raw/8472618112abcbd45acbcdc58436aff4233c23f7/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 918166240
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 15,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 197591040
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-02-16",
      "downloads": 122305,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "da33c16fa4440f831149fcf53b98a22bc07785e5",
      "configUrl": "https://huggingface.co/Qwen/Qwen3.5-397B-A17B/raw/8472618112abcbd45acbcdc58436aff4233c23f7/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 244093630912,
          "files": [
            "Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00001-of-00006.gguf",
            "Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00002-of-00006.gguf",
            "Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00003-of-00006.gguf",
            "Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00004-of-00006.gguf",
            "Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00005-of-00006.gguf",
            "Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00006-of-00006.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 421507365824,
          "files": [
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00001-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00002-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00003-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00004-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00005-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00006-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00007-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00008-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00009-of-00010.gguf",
            "Q8_0/Qwen3.5-397B-A17B-Q8_0-00010-of-00010.gguf"
          ]
        }
      ],
      "lab": "Qwen",
      "familyId": "qwen-qwen-3-5",
      "familyName": "Qwen 3.5",
      "likes": 1561,
      "sourceDownloads": 391204,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.042891778833306686,
        "activeParameters": 17000000000,
        "source": "https://huggingface.co/Qwen/Qwen3.5-397B-A17B"
      },
      "edition": "qwen3.5-397b-a17b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/gemma-4-12b-it-GGUF",
      "id": "google-gemma-4-12b-it",
      "name": "gemma-4-12B",
      "sourceRepo": "google/gemma-4-12B-it",
      "parameters": 11907350576,
      "layers": 48,
      "kvHeads": 8,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/google/gemma-4-12B-it/raw/707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 175115840
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 40,
            "bytesPerToken": 8192,
            "window": 1024
          },
          {
            "layers": 8,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-05-23",
      "downloads": 1520187,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "fc034cfff751157913579611efad8462ac1be606",
      "configUrl": "https://huggingface.co/google/gemma-4-12B-it/raw/707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 7121861440,
          "files": [
            "gemma-4-12b-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 12669647680,
          "files": [
            "gemma-4-12b-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-4",
      "familyName": "Gemma 4",
      "likes": 1629,
      "sourceDownloads": 1905338,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "gemma-4-12b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/gemma-4-26B-A4B-it-GGUF",
      "id": "google-gemma-4-26b-a4b-it",
      "name": "gemma-4-26B-A4B",
      "sourceRepo": "google/gemma-4-26B-A4B-it",
      "parameters": 25233142046,
      "layers": 30,
      "kvHeads": 8,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/google/gemma-4-26B-A4B-it/raw/4d7ae4984b7db7de8f8457170b3f1a419ee76d52/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 1193058784
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 25,
            "bytesPerToken": 8192,
            "window": 1024
          },
          {
            "layers": 5,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-03-11",
      "downloads": 514002,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "c099eb48e663fd284577b04978a94ffccb261841",
      "configUrl": "https://huggingface.co/google/gemma-4-26B-A4B-it/raw/4d7ae4984b7db7de8f8457170b3f1a419ee76d52/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 16947541728,
          "files": [
            "gemma-4-26B-A4B-it-UD-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 26859861728,
          "files": [
            "gemma-4-26B-A4B-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-4",
      "familyName": "Gemma 4",
      "likes": 1584,
      "sourceDownloads": 12963043,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.15852167727300875,
        "activeParameters": 4000000000,
        "source": "https://huggingface.co/google/gemma-4-26B-A4B-it"
      },
      "edition": "gemma-4-26b-a4b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/gemma-4-31B-it-GGUF",
      "id": "google-gemma-4-31b-it",
      "name": "gemma-4-31B",
      "sourceRepo": "google/gemma-4-31B-it",
      "parameters": 30697345596,
      "layers": 60,
      "kvHeads": 16,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/google/gemma-4-31B-it/raw/842da3794eaa0b77d5f08bae87a17459d91ff475/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 1198957024
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 50,
            "bytesPerToken": 16384,
            "window": 1024
          },
          {
            "layers": 10,
            "bytesPerToken": 8192
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-03-11",
      "downloads": 321798,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "c1ac76e99d5513b141e8adde7288b85c3f9c32ec",
      "configUrl": "https://huggingface.co/google/gemma-4-31B-it/raw/842da3794eaa0b77d5f08bae87a17459d91ff475/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 18323733440,
          "files": [
            "gemma-4-31B-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 32635677632,
          "files": [
            "gemma-4-31B-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-4",
      "familyName": "Gemma 4",
      "likes": 4005,
      "sourceDownloads": 9875118,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "gemma-4-31b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/gemma-4-E2B-it-GGUF",
      "id": "google-gemma-4-e2b-it",
      "name": "gemma-4-E2B",
      "sourceRepo": "google/gemma-4-E2B-it",
      "parameters": 4647450147,
      "layers": 35,
      "kvHeads": 1,
      "headDim": 256,
      "context": 131072,
      "contextSource": "https://huggingface.co/google/gemma-4-E2B-it/raw/3e22461f65e89153144f8adb70e3b8c2cc9845a7/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 985654080
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 28,
            "bytesPerToken": 1024,
            "window": 512
          },
          {
            "layers": 7,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-03-02",
      "downloads": 385562,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "0314792d7f1f7e229411f620751375812bb9faf2",
      "configUrl": "https://huggingface.co/google/gemma-4-E2B-it/raw/3e22461f65e89153144f8adb70e3b8c2cc9845a7/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 3106738272,
          "files": [
            "gemma-4-E2B-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 5048352864,
          "files": [
            "gemma-4-E2B-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-4",
      "familyName": "Gemma 4",
      "likes": 999,
      "sourceDownloads": 3043721,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "gemma-4-e2b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/gemma-4-E4B-it-GGUF",
      "id": "google-gemma-4-e4b-it",
      "name": "gemma-4-E4B",
      "sourceRepo": "google/gemma-4-E4B-it",
      "parameters": 7518069290,
      "layers": 42,
      "kvHeads": 2,
      "headDim": 256,
      "context": 131072,
      "contextSource": "https://huggingface.co/google/gemma-4-E4B-it/raw/ee0ef6023621cff504d758262d4e04895a5af4a2/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 990372672
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 35,
            "bytesPerToken": 2048,
            "window": 512
          },
          {
            "layers": 7,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-03-02",
      "downloads": 615083,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "bfc15c382204943c3a8fff0c750b94ae2364d7a3",
      "configUrl": "https://huggingface.co/google/gemma-4-E4B-it/raw/ee0ef6023621cff504d758262d4e04895a5af4a2/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 4977171584,
          "files": [
            "gemma-4-E4B-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 8192953472,
          "files": [
            "gemma-4-E4B-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-4",
      "familyName": "Gemma 4",
      "likes": 1635,
      "sourceDownloads": 4401295,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "gemma-4-e4b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "lab": "Google",
      "familyId": "google-gemma-3",
      "familyName": "Gemma 3",
      "edition": "gemma-3-270m-it",
      "sourceRepo": "google/gemma-3-270m-it",
      "released": "2025-07-30",
      "repo": "unsloth/gemma-3-270m-it-GGUF",
      "id": "google-gemma-3-270m-it",
      "name": "gemma-3-270m-it",
      "configRepo": "unsloth/gemma-3-270m-it",
      "parameters": 268098176,
      "layers": 18,
      "kvHeads": 1,
      "headDim": 256,
      "context": 32768,
      "contextSource": "https://huggingface.co/unsloth/gemma-3-270m-it/raw/23cf460f6bb16954176b3ddcc8d4f250501458a9/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 15,
            "bytesPerToken": 1024,
            "window": 512
          },
          {
            "layers": 3,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "gemma",
      "downloads": 70332,
      "likes": 648,
      "sourceDownloads": 70479,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "c90975dbd40c0c7b275fefaae758c3415c906238",
      "configUrl": "https://huggingface.co/unsloth/gemma-3-270m-it/raw/23cf460f6bb16954176b3ddcc8d4f250501458a9/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 253115424,
          "files": [
            "gemma-3-270m-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 291546144,
          "files": [
            "gemma-3-270m-it-Q8_0.gguf"
          ]
        }
      ],
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "lab": "Google",
      "familyId": "google-gemma-3",
      "familyName": "Gemma 3",
      "edition": "gemma-3-1b-it",
      "sourceRepo": "google/gemma-3-1b-it",
      "released": "2025-03-10",
      "repo": "unsloth/gemma-3-1b-it-GGUF",
      "id": "google-gemma-3-1b-it",
      "name": "gemma-3-1b-it",
      "configRepo": "unsloth/gemma-3-1b-it",
      "parameters": 999885952,
      "layers": 26,
      "kvHeads": 1,
      "headDim": 256,
      "context": 32768,
      "contextSource": "https://huggingface.co/unsloth/gemma-3-1b-it/raw/5b11413a10db4e486ef16a20101fd028f8f2499c/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 22,
            "bytesPerToken": 1024,
            "window": 512
          },
          {
            "layers": 4,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "gemma",
      "downloads": 45179,
      "likes": 1200,
      "sourceDownloads": 3342187,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "f0b45be0aac41bd6a100a4b5734cad5f67255bfb",
      "configUrl": "https://huggingface.co/unsloth/gemma-3-1b-it/raw/5b11413a10db4e486ef16a20101fd028f8f2499c/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 806058272,
          "files": [
            "gemma-3-1b-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 1069306400,
          "files": [
            "gemma-3-1b-it-Q8_0.gguf"
          ]
        }
      ],
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "sourceRepo": "google/gemma-3-12b-it",
      "repo": "unsloth/gemma-3-12b-it-GGUF",
      "id": "google-gemma-3-12b-it",
      "name": "gemma-3-12b",
      "configRepo": "unsloth/gemma-3-12b-it",
      "parameters": 11766034176,
      "layers": 48,
      "kvHeads": 8,
      "headDim": 256,
      "context": 131072,
      "contextSource": "https://huggingface.co/unsloth/gemma-3-12b-it/raw/9478e665381f42974aa06177b019352fb6291876/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 854200448
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 40,
            "bytesPerToken": 8192,
            "window": 1024
          },
          {
            "layers": 8,
            "bytesPerToken": 8192
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "gemma",
      "released": "2025-03-01",
      "downloads": 65479,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "d15e4c7dc21dc55d56bf8549db57a71ad8a2a35d",
      "configUrl": "https://huggingface.co/unsloth/gemma-3-12b-it/raw/9478e665381f42974aa06177b019352fb6291876/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 7300778336,
          "files": [
            "gemma-3-12b-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 12510212576,
          "files": [
            "gemma-3-12b-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-3",
      "familyName": "Gemma 3",
      "likes": 847,
      "sourceDownloads": 491997,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "gemma-3-12b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "sourceRepo": "google/gemma-3-27b-it",
      "repo": "unsloth/gemma-3-27b-it-GGUF",
      "id": "google-gemma-3-27b-it",
      "name": "gemma-3-27b",
      "configRepo": "unsloth/gemma-3-27b-it",
      "parameters": 27009346304,
      "layers": 62,
      "kvHeads": 16,
      "headDim": 128,
      "context": 131072,
      "contextSource": "https://huggingface.co/unsloth/gemma-3-27b-it/raw/7a5a3053dbd5d1d58e48159e87b9df2fc545a49a/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 857739392
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 52,
            "bytesPerToken": 8192,
            "window": 1024
          },
          {
            "layers": 10,
            "bytesPerToken": 8192
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "gemma",
      "released": "2025-03-01",
      "downloads": 35242,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "7cd0121f2530b00e42c4df952d4cad4418c0b3c1",
      "configUrl": "https://huggingface.co/unsloth/gemma-3-27b-it/raw/7a5a3053dbd5d1d58e48159e87b9df2fc545a49a/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 16546688736,
          "files": [
            "gemma-3-27b-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 28707972192,
          "files": [
            "gemma-3-27b-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-3",
      "familyName": "Gemma 3",
      "likes": 2044,
      "sourceDownloads": 417382,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "gemma-3-27b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "sourceRepo": "google/gemma-3-4b-it",
      "repo": "unsloth/gemma-3-4b-it-GGUF",
      "id": "google-gemma-3-4b-it",
      "name": "gemma-3-4b",
      "configRepo": "unsloth/gemma-3-4b-it",
      "parameters": 3880263168,
      "layers": 34,
      "kvHeads": 4,
      "headDim": 256,
      "context": 131072,
      "contextSource": "https://huggingface.co/unsloth/gemma-3-4b-it/raw/bf46152c47f5dd20b896357cb51abc4c03b8ee8c/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 851251328
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 29,
            "bytesPerToken": 4096,
            "window": 1024
          },
          {
            "layers": 5,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "gemma",
      "released": "2025-02-20",
      "downloads": 91803,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "5a3566e716d80f709ed7b79817eaf7733d2a1fce",
      "configUrl": "https://huggingface.co/unsloth/gemma-3-4b-it/raw/bf46152c47f5dd20b896357cb51abc4c03b8ee8c/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 2489894016,
          "files": [
            "gemma-3-4b-it-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 4130402176,
          "files": [
            "gemma-3-4b-it-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Google",
      "familyId": "google-gemma-3",
      "familyName": "Gemma 3",
      "likes": 1549,
      "sourceDownloads": 1258354,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "gemma-3-4b-it",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "vcruz305/DeepSeek-V4.1-Flash-GGUF",
      "id": "deepseek-ai-deepseek-v4-1-flash",
      "name": "DeepSeek-V4.1-Flash",
      "sourceRepo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "parameters": 748494684784,
      "layers": 40,
      "kvHeads": 1,
      "headDim": 512,
      "context": 1048576,
      "contextSource": "https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/dba1be0a40aa45a94ad051997016db3960a90277/config.json",
      "vision": true,
      "projector": null,
      "memory": {
        "kind": "unverified",
        "reason": "This model uses a custom cache; check memory requirements with its supported runner."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "mit",
      "released": "2026-09-10",
      "downloads": 44573,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "c4a085541cb53f67ee5e57b63d255e80cef286e7",
      "configUrl": "https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/2cba9e42aa026125f3ed06c6d98c1db82f7ca027/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 444737454720,
          "files": [
            "DeepSeek-V4.1-Flash-Q4_K_M-00001-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00002-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00003-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00004-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00005-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00006-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00007-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00008-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00009-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00010-of-00011.gguf",
            "DeepSeek-V4.1-Flash-Q4_K_M-00011-of-00011.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 507954227008,
          "files": [
            "DeepSeek-V4.1-Flash-Q8_0-00001-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00002-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00003-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00004-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00005-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00006-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00007-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00008-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00009-of-00010.gguf",
            "DeepSeek-V4.1-Flash-Q8_0-00010-of-00010.gguf"
          ]
        }
      ],
      "lab": "DeepSeek",
      "familyId": "deepseek-deepseek-v4-1-flash",
      "familyName": "DeepSeek V4.1 Flash",
      "likes": 4023,
      "sourceDownloads": 787841,
      "throughput": {
        "kind": "unverified"
      },
      "edition": "deepseek-v4.1-flash",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/DeepSeek-V4-Flash-Vision-Exp-GGUF",
      "id": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "name": "DeepSeek-V4-Flash-Vision-Exp",
      "sourceRepo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "parameters": 284334578519,
      "layers": 43,
      "kvHeads": 1,
      "headDim": 512,
      "context": 1048576,
      "contextSource": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-Vision-Exp/raw/6821d6ad3681a4b137b066b76094fa82ebd0a380/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 933258400
      },
      "memory": {
        "kind": "unverified",
        "reason": "This model uses a custom cache; check memory requirements with its supported runner."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "mit",
      "released": "2026-08-31",
      "downloads": 29907,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "87ce15810931a270f165e5e50439cd446f6b4ba5",
      "configUrl": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-Vision-Exp/raw/6821d6ad3681a4b137b066b76094fa82ebd0a380/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 155095288672,
          "files": [
            "UD-Q4_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q4_K_XL-00001-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q4_K_XL-00002-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q4_K_XL-00003-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q4_K_XL-00004-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q4_K_XL-00005-of-00005.gguf"
          ]
        },
        {
          "quant": "UD-Q8_K_XL",
          "precision": 8,
          "bytes": 161869663072,
          "files": [
            "UD-Q8_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q8_K_XL-00001-of-00005.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q8_K_XL-00002-of-00005.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q8_K_XL-00003-of-00005.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q8_K_XL-00004-of-00005.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Flash-Vision-Exp-UD-Q8_K_XL-00005-of-00005.gguf"
          ]
        }
      ],
      "lab": "DeepSeek",
      "familyId": "deepseek-deepseek-v4-flash-vision",
      "familyName": "DeepSeek V4 Flash Vision",
      "likes": 930,
      "sourceDownloads": 915320,
      "throughput": {
        "kind": "unverified"
      },
      "edition": "deepseek-v4-flash-vision-exp",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/DeepSeek-V4-Pro-0813-GGUF",
      "id": "deepseek-ai-deepseek-v4-pro-0813",
      "name": "DeepSeek-V4-Pro-0813",
      "sourceRepo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "parameters": 1572999528803,
      "layers": 61,
      "kvHeads": 1,
      "headDim": 512,
      "context": 1048576,
      "contextSource": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/raw/72e1d3230f6c080a530b0a1d46f8eb4602340597/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "unverified",
        "reason": "This model uses a custom cache; check memory requirements with its supported runner."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "mit",
      "released": "2026-08-13",
      "downloads": 117206,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "6d053616d152293da72569f56b861466f10ace7d",
      "configUrl": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/raw/72e1d3230f6c080a530b0a1d46f8eb4602340597/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 849683927055,
          "files": [
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00001-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00002-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00003-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00004-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00005-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00006-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00007-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00008-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00009-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00010-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00011-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00012-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00013-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00014-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00015-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00016-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00017-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00018-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00019-of-00020.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00020-of-00020.gguf"
          ]
        },
        {
          "quant": "UD-Q8_K_XL",
          "precision": 8,
          "bytes": 873445601295,
          "files": [
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00001-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00002-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00003-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00004-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00005-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00006-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00007-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00008-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00009-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00010-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00011-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00012-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00013-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00014-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00015-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00016-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00017-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00018-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00019-of-00020.gguf",
            "UD-Q8_K_XL/DeepSeek-V4-Pro-0813-UD-Q8_K_XL-00020-of-00020.gguf"
          ]
        }
      ],
      "lab": "DeepSeek",
      "familyId": "deepseek-deepseek-v4-pro",
      "familyName": "DeepSeek V4 Pro",
      "likes": 854,
      "sourceDownloads": 95489,
      "throughput": {
        "kind": "unverified"
      },
      "edition": "deepseek-v4-pro",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "lab": "DeepSeek",
      "familyId": "deepseek-deepseek-v4-flash",
      "familyName": "DeepSeek V4 Flash",
      "edition": "deepseek-v4-flash",
      "sourceRepo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "released": "2026-07-31",
      "repo": "unsloth/DeepSeek-V4-Flash-0731-GGUF",
      "id": "deepseek-ai-deepseek-v4-flash-0731",
      "name": "DeepSeek-V4-Flash-0731",
      "parameters": 284334567511,
      "layers": 43,
      "kvHeads": 1,
      "headDim": 512,
      "context": 1048576,
      "contextSource": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731/raw/7872f01b1d1fe23eabc4c98b48bffcef5a386062/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "unverified",
        "reason": "This model uses a custom cache; check memory requirements with its supported runner."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "mit",
      "downloads": 152816,
      "likes": 4012,
      "sourceDownloads": 4481923,
      "throughput": {
        "kind": "unverified"
      },
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "fbbb5b93fb787c21338159b0af3318bb3f4d9768",
      "configUrl": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731/raw/7872f01b1d1fe23eabc4c98b48bffcef5a386062/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 155095241120,
          "files": [
            "UD-Q4_K_XL/DeepSeek-V4-Flash-0731-UD-Q4_K_XL-00001-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-0731-UD-Q4_K_XL-00002-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-0731-UD-Q4_K_XL-00003-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-0731-UD-Q4_K_XL-00004-of-00005.gguf",
            "UD-Q4_K_XL/DeepSeek-V4-Flash-0731-UD-Q4_K_XL-00005-of-00005.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 10896057440,
          "files": [
            "dspark-DeepSeek-V4-Flash-0731-Q8_0.gguf"
          ]
        }
      ],
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/GLM-5.3-Flash-GGUF",
      "id": "zai-org-glm-5-3-flash",
      "name": "GLM-5.3-Flash",
      "sourceRepo": "zai-org/GLM-5.3-Flash",
      "parameters": 320759404382,
      "layers": 45,
      "kvHeads": 64,
      "headDim": 64,
      "context": 1048576,
      "contextSource": "https://huggingface.co/zai-org/GLM-5.3-Flash/raw/eb9eb208eb0d988989d07a6a12d0fdeb5f52574a/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 1128047200
      },
      "memory": {
        "kind": "unverified",
        "reason": "This model uses a custom cache; check memory requirements with its supported runner."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "mit",
      "released": "2026-08-25",
      "downloads": 1036417,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "621d456e93e926e4b52f85cff5f634358c1828f9",
      "configUrl": "https://huggingface.co/zai-org/GLM-5.3-Flash/raw/eb9eb208eb0d988989d07a6a12d0fdeb5f52574a/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 199707321347,
          "files": [
            "UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00001-of-00006.gguf",
            "UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00002-of-00006.gguf",
            "UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00003-of-00006.gguf",
            "UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00004-of-00006.gguf",
            "UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00005-of-00006.gguf",
            "UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00006-of-00006.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 340981966112,
          "files": [
            "Q8_0/GLM-5.3-Flash-Q8_0-00001-of-00008.gguf",
            "Q8_0/GLM-5.3-Flash-Q8_0-00002-of-00008.gguf",
            "Q8_0/GLM-5.3-Flash-Q8_0-00003-of-00008.gguf",
            "Q8_0/GLM-5.3-Flash-Q8_0-00004-of-00008.gguf",
            "Q8_0/GLM-5.3-Flash-Q8_0-00005-of-00008.gguf",
            "Q8_0/GLM-5.3-Flash-Q8_0-00006-of-00008.gguf",
            "Q8_0/GLM-5.3-Flash-Q8_0-00007-of-00008.gguf",
            "Q8_0/GLM-5.3-Flash-Q8_0-00008-of-00008.gguf"
          ]
        }
      ],
      "lab": "Z.ai",
      "familyId": "z-ai-glm-5-3",
      "familyName": "GLM 5.3",
      "likes": 2668,
      "sourceDownloads": 5428853,
      "throughput": {
        "kind": "unverified"
      },
      "edition": "glm-5.3-flash",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/GLM-5.3-GGUF",
      "id": "zai-org-glm-5-3",
      "name": "GLM-5.3",
      "sourceRepo": "zai-org/GLM-5.3",
      "parameters": 753864139008,
      "layers": 78,
      "kvHeads": 64,
      "headDim": 192,
      "context": 1048576,
      "contextSource": "https://huggingface.co/zai-org/GLM-5.3/raw/aca966e4e02791568aa6a4ced368624b3d897f42/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "latent",
        "groups": [
          {
            "layers": 78,
            "bytesPerToken": 1412
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner that stores the compressed latent KV cache. Expanded caches need more memory."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-08-25",
      "downloads": 560708,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "346b3591c7f28d1a23716f97a065ecf12ec14771",
      "configUrl": "https://huggingface.co/zai-org/GLM-5.3/raw/aca966e4e02791568aa6a4ced368624b3d897f42/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 467289116837,
          "files": [
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00001-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00002-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00003-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00004-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00005-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00006-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00007-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00008-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00009-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00010-of-00011.gguf",
            "UD-Q4_K_XL/GLM-5.3-UD-Q4_K_XL-00011-of-00011.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 801357677216,
          "files": [
            "Q8_0/GLM-5.3-Q8_0-00001-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00002-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00003-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00004-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00005-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00006-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00007-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00008-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00009-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00010-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00011-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00012-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00013-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00014-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00015-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00016-of-00017.gguf",
            "Q8_0/GLM-5.3-Q8_0-00017-of-00017.gguf"
          ]
        }
      ],
      "lab": "Z.ai",
      "familyId": "z-ai-glm-5-3",
      "familyName": "GLM 5.3",
      "likes": 2063,
      "sourceDownloads": 1402948,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.06862993838131218,
        "source": "https://huggingface.co/zai-org/GLM-5.3"
      },
      "edition": "glm-5.3",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/GLM-5.2-GGUF",
      "id": "zai-org-glm-5-2",
      "name": "GLM-5.2",
      "sourceRepo": "zai-org/GLM-5.2",
      "parameters": 753864139008,
      "layers": 78,
      "kvHeads": 64,
      "headDim": 192,
      "context": 1048576,
      "contextSource": "https://huggingface.co/zai-org/GLM-5.2/raw/cf457fa734ab149ffef225f80893eb38c6ff5cdc/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "latent",
        "groups": [
          {
            "layers": 78,
            "bytesPerToken": 1412
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner that stores the compressed latent KV cache. Expanded caches need more memory."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "mit",
      "released": "2026-06-16",
      "downloads": 320942,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "abc55e72527792c6e77069c99b4cb7de16fa9f23",
      "configUrl": "https://huggingface.co/zai-org/GLM-5.2/raw/cf457fa734ab149ffef225f80893eb38c6ff5cdc/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 465825525088,
          "files": [
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00001-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00002-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00003-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00004-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00005-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00006-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00007-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00008-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00009-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00010-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00011-of-00011.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 801357672256,
          "files": [
            "Q8_0/GLM-5.2-Q8_0-00001-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00002-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00003-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00004-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00005-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00006-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00007-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00008-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00009-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00010-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00011-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00012-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00013-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00014-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00015-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00016-of-00017.gguf",
            "Q8_0/GLM-5.2-Q8_0-00017-of-00017.gguf"
          ]
        }
      ],
      "lab": "Z.ai",
      "familyId": "z-ai-glm-5-2",
      "familyName": "GLM 5.2",
      "likes": 5143,
      "sourceDownloads": 705839,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.06862993838131218,
        "source": "https://huggingface.co/zai-org/GLM-5.2"
      },
      "edition": "glm-5.2",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/GLM-5.1-GGUF",
      "id": "zai-org-glm-5-1",
      "name": "GLM-5.1",
      "sourceRepo": "zai-org/GLM-5.1",
      "parameters": 753864139008,
      "layers": 78,
      "kvHeads": 64,
      "headDim": 64,
      "context": 202752,
      "contextSource": "https://huggingface.co/zai-org/GLM-5.1/raw/26e1bd6e011feb778d25ae34b09b07074139d92d/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "latent",
        "groups": [
          {
            "layers": 78,
            "bytesPerToken": 1412
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner that stores the compressed latent KV cache. Expanded caches need more memory."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "mit",
      "released": "2026-04-03",
      "downloads": 3729,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "3238253553497e969f3144fda297dac98b99dbbe",
      "configUrl": "https://huggingface.co/zai-org/GLM-5.1/raw/26e1bd6e011feb778d25ae34b09b07074139d92d/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 464504318304,
          "files": [
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00001-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00002-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00003-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00004-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00005-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00006-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00007-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00008-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00009-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00010-of-00011.gguf",
            "UD-Q4_K_M/GLM-5.1-UD-Q4_K_M-00011-of-00011.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 801344965472,
          "files": [
            "Q8_0/GLM-5.1-Q8_0-00001-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00002-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00003-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00004-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00005-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00006-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00007-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00008-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00009-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00010-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00011-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00012-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00013-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00014-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00015-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00016-of-00017.gguf",
            "Q8_0/GLM-5.1-Q8_0-00017-of-00017.gguf"
          ]
        }
      ],
      "lab": "Z.ai",
      "familyId": "z-ai-glm-5-1",
      "familyName": "GLM 5.1",
      "likes": 1840,
      "sourceDownloads": 267129,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.06862993838131218,
        "source": "https://huggingface.co/zai-org/GLM-5.1"
      },
      "edition": "glm-5.1",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Mistral-Medium-3.5-128B-GGUF",
      "id": "mistralai-mistral-medium-3-5-128b",
      "name": "Mistral-Medium-3.5-128B",
      "sourceRepo": "mistralai/Mistral-Medium-3.5-128B",
      "parameters": 125025988608,
      "layers": 88,
      "kvHeads": 8,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/mistralai/Mistral-Medium-3.5-128B/raw/22b2b868a15677cfa6061277ed2f653d1349a9ab/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 5356846528
      },
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 88,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "other",
      "released": "2026-03-31",
      "downloads": 12052,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "c8f5b1477e1b22cd2d819157d450f001f7047298",
      "configUrl": "https://huggingface.co/mistralai/Mistral-Medium-3.5-128B/raw/22b2b868a15677cfa6061277ed2f653d1349a9ab/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 74897139136,
          "files": [
            "Q4_K_M/Mistral-Medium-3.5-128B-Q4_K_M-00001-of-00003.gguf",
            "Q4_K_M/Mistral-Medium-3.5-128B-Q4_K_M-00002-of-00003.gguf",
            "Q4_K_M/Mistral-Medium-3.5-128B-Q4_K_M-00003-of-00003.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 132854425088,
          "files": [
            "Q8_0/Mistral-Medium-3.5-128B-Q8_0-00001-of-00004.gguf",
            "Q8_0/Mistral-Medium-3.5-128B-Q8_0-00002-of-00004.gguf",
            "Q8_0/Mistral-Medium-3.5-128B-Q8_0-00003-of-00004.gguf",
            "Q8_0/Mistral-Medium-3.5-128B-Q8_0-00004-of-00004.gguf"
          ]
        }
      ],
      "lab": "Mistral",
      "familyId": "mistral-mistral-medium-3-5",
      "familyName": "Mistral Medium 3.5",
      "likes": 461,
      "sourceDownloads": 118113,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "mistral-medium-3.5-128b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Mistral-Small-4-119B-2603-GGUF",
      "id": "mistralai-mistral-small-4-119b-2603",
      "name": "Mistral-Small-4-119B-2603",
      "sourceRepo": "mistralai/Mistral-Small-4-119B-2603",
      "parameters": 118972826624,
      "layers": 36,
      "kvHeads": 32,
      "headDim": 128,
      "context": 1048576,
      "contextSource": "https://huggingface.co/mistralai/Mistral-Small-4-119B-2603/raw/a11f36bebf709121056b1dbcc943d1c6afbe494d/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 857078496
      },
      "memory": {
        "kind": "latent",
        "groups": [
          {
            "layers": 36,
            "bytesPerToken": 640
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner that stores the compressed latent KV cache. Expanded caches need more memory."
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-01-23",
      "downloads": 14560,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "bd93c721735aa32c035c0f19e738cb3371fd56ff",
      "configUrl": "https://huggingface.co/mistralai/Mistral-Small-4-119B-2603/raw/a11f36bebf709121056b1dbcc943d1c6afbe494d/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 73763180544,
          "files": [
            "UD-Q4_K_M/Mistral-Small-4-119B-2603-UD-Q4_K_M-00001-of-00003.gguf",
            "UD-Q4_K_M/Mistral-Small-4-119B-2603-UD-Q4_K_M-00002-of-00003.gguf",
            "UD-Q4_K_M/Mistral-Small-4-119B-2603-UD-Q4_K_M-00003-of-00003.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 126472999072,
          "files": [
            "Q8_0/Mistral-Small-4-119B-2603-Q8_0-00001-of-00004.gguf",
            "Q8_0/Mistral-Small-4-119B-2603-Q8_0-00002-of-00004.gguf",
            "Q8_0/Mistral-Small-4-119B-2603-Q8_0-00003-of-00004.gguf",
            "Q8_0/Mistral-Small-4-119B-2603-Q8_0-00004-of-00004.gguf"
          ]
        }
      ],
      "lab": "Mistral",
      "familyId": "mistral-mistral-small-4",
      "familyName": "Mistral Small 4",
      "likes": 430,
      "sourceDownloads": 54290,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.0557487661359979,
        "source": "https://huggingface.co/mistralai/Mistral-Small-4-119B-2603"
      },
      "edition": "mistral-small-4-119b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF",
      "id": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "name": "Nemotron-3.5-Lightning-30B-A3B",
      "sourceRepo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "parameters": 32913266240,
      "layers": 52,
      "kvHeads": 2,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16/raw/a9904d24bcc1d289a1950fa9d2b978c47cf903b9/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 50495488
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-08-01",
      "downloads": 66349,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "f2d3fe3694501008786e81e5f20360cbf715496a",
      "configUrl": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16/raw/a9904d24bcc1d289a1950fa9d2b978c47cf903b9/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 25266255936,
          "files": [
            "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-UD-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 35004643392,
          "files": [
            "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "NVIDIA",
      "familyId": "nvidia-nemotron-3-5",
      "familyName": "Nemotron 3.5",
      "likes": 226,
      "sourceDownloads": 530047,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.09114865653637419,
        "activeParameters": 3000000000,
        "source": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16"
      },
      "edition": "nvidia-nemotron-3.5-lightning-30b-a3b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF",
      "id": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "name": "Nemotron-3-Ultra-550B-A55B",
      "sourceRepo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "parameters": 549308993536,
      "kvHeads": 2,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16/raw/77df655d5e9f8362164ed14dd8b48f8bce657498/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 12,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 416808960
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-06-03",
      "downloads": 7375,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "2fb7d5b3f4eae7aedb18b4839b6a6300111e46f6",
      "configUrl": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16/raw/77df655d5e9f8362164ed14dd8b48f8bce657498/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 359225442240,
          "files": [
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00001-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00002-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00003-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00004-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00005-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00006-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00007-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00008-of-00009.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_M-00009-of-00009.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 584258241056,
          "files": [
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00001-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00002-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00003-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00004-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00005-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00006-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00007-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00008-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00009-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00010-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00011-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00012-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00013-of-00014.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Ultra-550B-A55B-Q8_0-00014-of-00014.gguf"
          ]
        }
      ],
      "lab": "NVIDIA",
      "familyId": "nvidia-nemotron-3",
      "familyName": "Nemotron 3",
      "layers": 108,
      "likes": 358,
      "sourceDownloads": 357957,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.10012579558538663,
        "activeParameters": 55000000000,
        "source": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16"
      },
      "edition": "nvidia-nemotron-3-ultra-550b-a55b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF",
      "id": "nvidia-nemotron-3-nano-omni-30b-a3b-reasoning-bf16",
      "name": "Nemotron-3-Nano-Omni-30B-A3B-Reasoning",
      "sourceRepo": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16",
      "parameters": 31577940288,
      "layers": 52,
      "kvHeads": 2,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16/raw/e5e9932441de940c9a62185c870ea5bcd4cd24e2/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 1587540224
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 50495488
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "other",
      "released": "2026-04-20",
      "downloads": 12483,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "571758804835f56154718683f5c0e388b7d0fef9",
      "configUrl": "https://huggingface.co/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16/raw/e5e9932441de940c9a62185c870ea5bcd4cd24e2/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 23887023552,
          "files": [
            "NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-UD-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 33585499584,
          "files": [
            "NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-Q8_0.gguf"
          ]
        }
      ],
      "lab": "NVIDIA",
      "familyId": "nvidia-nemotron-3-nano-omni",
      "familyName": "Nemotron 3 Nano Omni",
      "likes": 436,
      "sourceDownloads": 362114,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.09500302973022076,
        "activeParameters": 3000000000,
        "source": "https://huggingface.co/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16"
      },
      "edition": "nemotron-3-nano-omni-30b-a3b-reasoning",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF",
      "id": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "name": "Nemotron-3-Nano-4B",
      "sourceRepo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "parameters": 3973556832,
      "layers": 42,
      "kvHeads": 8,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16/raw/dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 4,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 85843968
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-03-07",
      "downloads": 7509,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "8e81be55c5aa3d63bb82b6ceec62d50805d9e1bb",
      "configUrl": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16/raw/dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 2900295712,
          "files": [
            "NVIDIA-Nemotron-3-Nano-4B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 4233679008,
          "files": [
            "NVIDIA-Nemotron-3-Nano-4B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "NVIDIA",
      "familyId": "nvidia-nemotron-3",
      "familyName": "Nemotron 3",
      "likes": 126,
      "sourceDownloads": 3418737,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "nvidia-nemotron-3-nano-4b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF",
      "id": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "name": "Nemotron-3-Super-120B-A12B",
      "sourceRepo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "parameters": 120668707840,
      "layers": 88,
      "kvHeads": 2,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16/raw/2dc98e2afe4face0e4ce40972a915c45368bd34a/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 8,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 174325760
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-03-10",
      "downloads": 13409,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "036038fb30334a2d56a146c6f0d4871ab5edccbb",
      "configUrl": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16/raw/2dc98e2afe4face0e4ce40972a915c45368bd34a/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 82541168480,
          "files": [
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_M-00001-of-00003.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_M-00002-of-00003.gguf",
            "UD-Q4_K_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_M-00003-of-00003.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 128472991680,
          "files": [
            "Q8_0/NVIDIA-Nemotron-3-Super-120B-A12B-Q8_0-00001-of-00004.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Super-120B-A12B-Q8_0-00002-of-00004.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Super-120B-A12B-Q8_0-00003-of-00004.gguf",
            "Q8_0/NVIDIA-Nemotron-3-Super-120B-A12B-Q8_0-00004-of-00004.gguf"
          ]
        }
      ],
      "lab": "NVIDIA",
      "familyId": "nvidia-nemotron-3",
      "familyName": "Nemotron 3",
      "likes": 429,
      "sourceDownloads": 1193953,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.09944583160624652,
        "activeParameters": 12000000000,
        "source": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16"
      },
      "edition": "nvidia-nemotron-3-super-120b-a12b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Kimi-K3-GGUF",
      "id": "moonshotai-kimi-k3",
      "name": "Kimi-K3",
      "sourceRepo": "moonshotai/Kimi-K3",
      "parameters": 2779483135584,
      "layers": 93,
      "kvHeads": 96,
      "headDim": 74.66666666666667,
      "context": 1048576,
      "contextSource": "https://huggingface.co/moonshotai/Kimi-K3/raw/f831ab66814297da540d832a5235f8e904f29d06/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 903244832
      },
      "memory": {
        "kind": "unverified",
        "reason": "This model uses a custom cache; check memory requirements with its supported runner."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "other",
      "released": "2026-06-13",
      "downloads": 371886,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "a0836360ce58dfec088d966a97f2ddc8a606279b",
      "configUrl": "https://huggingface.co/moonshotai/Kimi-K3/raw/f831ab66814297da540d832a5235f8e904f29d06/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 1508668683104,
          "files": [
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00001-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00002-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00003-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00004-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00005-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00006-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00007-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00008-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00009-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00010-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00011-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00012-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00013-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00014-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00015-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00016-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00017-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00018-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00019-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00020-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00021-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00022-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00023-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00024-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00025-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00026-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00027-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00028-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00029-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00030-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00031-of-00032.gguf",
            "UD-Q4_K_XL/Kimi-K3-UD-Q4_K_XL-00032-of-00032.gguf"
          ]
        },
        {
          "quant": "UD-Q8_K_XL",
          "precision": 8,
          "bytes": 1561157884384,
          "files": [
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00001-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00002-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00003-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00004-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00005-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00006-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00007-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00008-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00009-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00010-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00011-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00012-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00013-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00014-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00015-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00016-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00017-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00018-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00019-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00020-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00021-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00022-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00023-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00024-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00025-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00026-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00027-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00028-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00029-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00030-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00031-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00032-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00033-of-00034.gguf",
            "UD-Q8_K_XL/Kimi-K3-UD-Q8_K_XL-00034-of-00034.gguf"
          ]
        }
      ],
      "lab": "Moonshot AI",
      "familyId": "moonshot-ai-kimi-k3",
      "familyName": "Kimi K3",
      "likes": 11579,
      "sourceDownloads": 1223892,
      "throughput": {
        "kind": "unverified"
      },
      "edition": "kimi-k3",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Kimi-K2.7-Code-GGUF",
      "id": "moonshotai-kimi-k2-7-code",
      "name": "Kimi-K2.7-Code",
      "sourceRepo": "moonshotai/Kimi-K2.7-Code",
      "parameters": 1026408232448,
      "layers": 61,
      "kvHeads": 64,
      "headDim": 112,
      "context": 262144,
      "contextSource": "https://huggingface.co/moonshotai/Kimi-K2.7-Code/raw/74797c9c62378b951a1f6fcf5c4631024e9b8bef/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 952572320
      },
      "memory": {
        "kind": "latent",
        "groups": [
          {
            "layers": 61,
            "bytesPerToken": 1152
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner that stores the compressed latent KV cache. Expanded caches need more memory."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "other",
      "released": "2026-06-11",
      "downloads": 334924,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "46352ca8dc32aa60f9754f5bd3fb778deeb9e430",
      "configUrl": "https://huggingface.co/moonshotai/Kimi-K2.7-Code/raw/74797c9c62378b951a1f6fcf5c4631024e9b8bef/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 583710875520,
          "files": [
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00001-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00002-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00003-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00004-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00005-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00006-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00007-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00008-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00009-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00010-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00011-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00012-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00013-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-00014-of-00014.gguf"
          ]
        },
        {
          "quant": "UD-Q8_K_XL",
          "precision": 8,
          "bytes": 594544652160,
          "files": [
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00001-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00002-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00003-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00004-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00005-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00006-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00007-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00008-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00009-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00010-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00011-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00012-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00013-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-00014-of-00014.gguf"
          ]
        }
      ],
      "lab": "Moonshot AI",
      "familyId": "moonshot-ai-kimi-k2-7-code",
      "familyName": "Kimi K2.7 Code",
      "likes": 1407,
      "sourceDownloads": 110745,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.03201601457309515,
        "source": "https://huggingface.co/moonshotai/Kimi-K2.7-Code"
      },
      "edition": "kimi-k2.7-code",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Kimi-K2.6-GGUF",
      "id": "moonshotai-kimi-k2-6",
      "name": "Kimi-K2.6",
      "sourceRepo": "moonshotai/Kimi-K2.6",
      "parameters": 1026408232448,
      "layers": 61,
      "kvHeads": 64,
      "headDim": 112,
      "context": 262144,
      "contextSource": "https://huggingface.co/moonshotai/Kimi-K2.6/raw/7eb5002f6aadc958aed6a9177b7ed26bb94011bb/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 952572480
      },
      "memory": {
        "kind": "latent",
        "groups": [
          {
            "layers": 61,
            "bytesPerToken": 1152
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner that stores the compressed latent KV cache. Expanded caches need more memory."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "other",
      "released": "2026-04-14",
      "downloads": 165784,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "47c7cab1e440dd9fcfc57c469e4737983408a6f2",
      "configUrl": "https://huggingface.co/moonshotai/Kimi-K2.6/raw/7eb5002f6aadc958aed6a9177b7ed26bb94011bb/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_XL",
          "precision": 4,
          "bytes": 583710875520,
          "files": [
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00001-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00002-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00003-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00004-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00005-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00006-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00007-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00008-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00009-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00010-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00011-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00012-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00013-of-00014.gguf",
            "UD-Q4_K_XL/Kimi-K2.6-UD-Q4_K_XL-00014-of-00014.gguf"
          ]
        },
        {
          "quant": "UD-Q8_K_XL",
          "precision": 8,
          "bytes": 594544652160,
          "files": [
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00001-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00002-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00003-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00004-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00005-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00006-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00007-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00008-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00009-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00010-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00011-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00012-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00013-of-00014.gguf",
            "UD-Q8_K_XL/Kimi-K2.6-UD-Q8_K_XL-00014-of-00014.gguf"
          ]
        }
      ],
      "lab": "Moonshot AI",
      "familyId": "moonshot-ai-kimi-k2-6",
      "familyName": "Kimi K2.6",
      "likes": 1612,
      "sourceDownloads": 472766,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.03201601457309515,
        "source": "https://huggingface.co/moonshotai/Kimi-K2.6"
      },
      "edition": "kimi-k2.6",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/MiniMax-M3-GGUF",
      "id": "minimaxai-minimax-m3",
      "name": "MiniMax-M3",
      "sourceRepo": "MiniMaxAI/MiniMax-M3",
      "parameters": 426174572928,
      "layers": 60,
      "kvHeads": 4,
      "headDim": 128,
      "context": 1048576,
      "contextSource": "https://huggingface.co/MiniMaxAI/MiniMax-M3/raw/f0e1c1e04d40177e4673a22097036854f536e9c0/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-MiniMax-M3-f16.gguf",
        "bytes": 1732284000
      },
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 60,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "other",
      "released": "2026-06-02",
      "downloads": 5446,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "2ee48e5fcf21a2522ee2fb9ffaa15592ec5498e6",
      "configUrl": "https://huggingface.co/MiniMaxAI/MiniMax-M3/raw/f0e1c1e04d40177e4673a22097036854f536e9c0/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 261282110272,
          "files": [
            "MiniMax-M3-Q4_K_M/MiniMax-M3-Q4_K_M-00001-of-00007.gguf",
            "MiniMax-M3-Q4_K_M/MiniMax-M3-Q4_K_M-00002-of-00007.gguf",
            "MiniMax-M3-Q4_K_M/MiniMax-M3-Q4_K_M-00003-of-00007.gguf",
            "MiniMax-M3-Q4_K_M/MiniMax-M3-Q4_K_M-00004-of-00007.gguf",
            "MiniMax-M3-Q4_K_M/MiniMax-M3-Q4_K_M-00005-of-00007.gguf",
            "MiniMax-M3-Q4_K_M/MiniMax-M3-Q4_K_M-00006-of-00007.gguf",
            "MiniMax-M3-Q4_K_M/MiniMax-M3-Q4_K_M-00007-of-00007.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 453611164768,
          "files": [
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00001-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00002-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00003-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00004-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00005-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00006-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00007-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00008-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00009-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00010-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00011-of-00012.gguf",
            "MiniMax-M3-Q8_0/MiniMax-M3-Q8_0-00012-of-00012.gguf"
          ]
        }
      ],
      "lab": "MiniMax",
      "familyId": "minimax-minimax-m3",
      "familyName": "MiniMax M3",
      "likes": 1557,
      "sourceDownloads": 177216,
      "throughput": {
        "kind": "unverified"
      },
      "edition": "minimax-m3",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/MiniMax-M2.7-GGUF",
      "id": "minimaxai-minimax-m2-7",
      "name": "MiniMax-M2.7",
      "sourceRepo": "MiniMaxAI/MiniMax-M2.7",
      "parameters": 228689764864,
      "layers": 62,
      "kvHeads": 8,
      "headDim": 128,
      "context": 204800,
      "contextSource": "https://huggingface.co/MiniMaxAI/MiniMax-M2.7/raw/d494266a4affc0d2995ba1fa35c8481cbd84294b/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 62,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-04-09",
      "downloads": 7055,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "d2a05ccf69491b03db0cc40b335aec14bdaf7198",
      "configUrl": "https://huggingface.co/MiniMaxAI/MiniMax-M2.7/raw/d494266a4affc0d2995ba1fa35c8481cbd84294b/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 140169905504,
          "files": [
            "UD-Q4_K_M/MiniMax-M2.7-UD-Q4_K_M-00001-of-00004.gguf",
            "UD-Q4_K_M/MiniMax-M2.7-UD-Q4_K_M-00002-of-00004.gguf",
            "UD-Q4_K_M/MiniMax-M2.7-UD-Q4_K_M-00003-of-00004.gguf",
            "UD-Q4_K_M/MiniMax-M2.7-UD-Q4_K_M-00004-of-00004.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 243136872992,
          "files": [
            "Q8_0/MiniMax-M2.7-Q8_0-00001-of-00006.gguf",
            "Q8_0/MiniMax-M2.7-Q8_0-00002-of-00006.gguf",
            "Q8_0/MiniMax-M2.7-Q8_0-00003-of-00006.gguf",
            "Q8_0/MiniMax-M2.7-Q8_0-00004-of-00006.gguf",
            "Q8_0/MiniMax-M2.7-Q8_0-00005-of-00006.gguf",
            "Q8_0/MiniMax-M2.7-Q8_0-00006-of-00006.gguf"
          ]
        }
      ],
      "lab": "MiniMax",
      "familyId": "minimax-minimax-m2-7",
      "familyName": "MiniMax M2.7",
      "likes": 1249,
      "sourceDownloads": 1073367,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.048233698148055656,
        "source": "https://huggingface.co/MiniMaxAI/MiniMax-M2.7"
      },
      "edition": "minimax-m2.7",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/MiniMax-M2.5-GGUF",
      "id": "minimaxai-minimax-m2-5",
      "name": "MiniMax-M2.5",
      "sourceRepo": "MiniMaxAI/MiniMax-M2.5",
      "parameters": 228689764864,
      "layers": 62,
      "kvHeads": 8,
      "headDim": 128,
      "context": 196608,
      "contextSource": "https://huggingface.co/MiniMaxAI/MiniMax-M2.5/raw/f710177d938eff80b684d42c5aa84b382612f21f/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 62,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "other",
      "released": "2026-02-12",
      "downloads": 5966,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "7c50dca0e5592483ad308ecffc876aecac725660",
      "configUrl": "https://huggingface.co/MiniMaxAI/MiniMax-M2.5/raw/f710177d938eff80b684d42c5aa84b382612f21f/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 138342385216,
          "files": [
            "Q4_K_M/MiniMax-M2.5-Q4_K_M-00001-of-00004.gguf",
            "Q4_K_M/MiniMax-M2.5-Q4_K_M-00002-of-00004.gguf",
            "Q4_K_M/MiniMax-M2.5-Q4_K_M-00003-of-00004.gguf",
            "Q4_K_M/MiniMax-M2.5-Q4_K_M-00004-of-00004.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 243136873280,
          "files": [
            "Q8_0/MiniMax-M2.5-Q8_0-00001-of-00006.gguf",
            "Q8_0/MiniMax-M2.5-Q8_0-00002-of-00006.gguf",
            "Q8_0/MiniMax-M2.5-Q8_0-00003-of-00006.gguf",
            "Q8_0/MiniMax-M2.5-Q8_0-00004-of-00006.gguf",
            "Q8_0/MiniMax-M2.5-Q8_0-00005-of-00006.gguf",
            "Q8_0/MiniMax-M2.5-Q8_0-00006-of-00006.gguf"
          ]
        }
      ],
      "lab": "MiniMax",
      "familyId": "minimax-minimax-m2-5",
      "familyName": "MiniMax M2.5",
      "likes": 1507,
      "sourceDownloads": 444793,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.048233698148055656,
        "source": "https://huggingface.co/MiniMaxAI/MiniMax-M2.5"
      },
      "edition": "minimax-m2.5",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/Step-3.7-Flash-GGUF",
      "id": "stepfun-ai-step-3-7-flash",
      "name": "Step-3.7-Flash",
      "sourceRepo": "stepfun-ai/Step-3.7-Flash",
      "parameters": 196956130432,
      "layers": 45,
      "kvHeads": 8,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/stepfun-ai/Step-3.7-Flash/raw/5f6244077ac62e04eec3f320501ff8c2b293373a/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 3972829408
      },
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 12,
            "bytesPerToken": 4096
          },
          {
            "layers": 33,
            "bytesPerToken": 4096,
            "window": 512
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-05-23",
      "downloads": 218782,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "8dd62fe8a98f3c8994160d0c521542560b558f37",
      "configUrl": "https://huggingface.co/stepfun-ai/Step-3.7-Flash/raw/5f6244077ac62e04eec3f320501ff8c2b293373a/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 122090426976,
          "files": [
            "UD-Q4_K_M/Step-3.7-Flash-UD-Q4_K_M-00001-of-00004.gguf",
            "UD-Q4_K_M/Step-3.7-Flash-UD-Q4_K_M-00002-of-00004.gguf",
            "UD-Q4_K_M/Step-3.7-Flash-UD-Q4_K_M-00003-of-00004.gguf",
            "UD-Q4_K_M/Step-3.7-Flash-UD-Q4_K_M-00004-of-00004.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 209417872192,
          "files": [
            "Q8_0/Step-3.7-Flash-Q8_0-00001-of-00006.gguf",
            "Q8_0/Step-3.7-Flash-Q8_0-00002-of-00006.gguf",
            "Q8_0/Step-3.7-Flash-Q8_0-00003-of-00006.gguf",
            "Q8_0/Step-3.7-Flash-Q8_0-00004-of-00006.gguf",
            "Q8_0/Step-3.7-Flash-Q8_0-00005-of-00006.gguf",
            "Q8_0/Step-3.7-Flash-Q8_0-00006-of-00006.gguf"
          ]
        }
      ],
      "lab": "StepFun",
      "familyId": "stepfun-step-3-7-flash",
      "familyName": "Step 3.7 Flash",
      "likes": 460,
      "sourceDownloads": 24803,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.060862914019011345,
        "source": "https://huggingface.co/stepfun-ai/Step-3.7-Flash"
      },
      "edition": "step-3.7-flash",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/stepfun-ai_Step-3.5-Flash-GGUF",
      "id": "stepfun-ai-step-3-5-flash",
      "name": "Step-3.5-Flash",
      "sourceRepo": "stepfun-ai/Step-3.5-Flash",
      "parameters": 196956130432,
      "layers": 45,
      "kvHeads": 8,
      "headDim": 128,
      "context": 262144,
      "contextSource": "https://huggingface.co/stepfun-ai/Step-3.5-Flash/raw/ab446a3de5e171ea341227e24bb1f090e1b771f7/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 12,
            "bytesPerToken": 4096
          },
          {
            "layers": 33,
            "bytesPerToken": 4096,
            "window": 512
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "apache-2.0",
      "released": "2026-02-01",
      "downloads": 1665,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "a2943ef56a0a240bdfb9a2feee97a775f612476d",
      "configUrl": "https://huggingface.co/stepfun-ai/Step-3.5-Flash/raw/ab446a3de5e171ea341227e24bb1f090e1b771f7/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 119937953440,
          "files": [
            "stepfun-ai_Step-3.5-Flash-Q4_K_M/stepfun-ai_Step-3.5-Flash-Q4_K_M-00001-of-00004.gguf",
            "stepfun-ai_Step-3.5-Flash-Q4_K_M/stepfun-ai_Step-3.5-Flash-Q4_K_M-00002-of-00004.gguf",
            "stepfun-ai_Step-3.5-Flash-Q4_K_M/stepfun-ai_Step-3.5-Flash-Q4_K_M-00003-of-00004.gguf",
            "stepfun-ai_Step-3.5-Flash-Q4_K_M/stepfun-ai_Step-3.5-Flash-Q4_K_M-00004-of-00004.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 209417870976,
          "files": [
            "stepfun-ai_Step-3.5-Flash-Q8_0/stepfun-ai_Step-3.5-Flash-Q8_0-00001-of-00006.gguf",
            "stepfun-ai_Step-3.5-Flash-Q8_0/stepfun-ai_Step-3.5-Flash-Q8_0-00002-of-00006.gguf",
            "stepfun-ai_Step-3.5-Flash-Q8_0/stepfun-ai_Step-3.5-Flash-Q8_0-00003-of-00006.gguf",
            "stepfun-ai_Step-3.5-Flash-Q8_0/stepfun-ai_Step-3.5-Flash-Q8_0-00004-of-00006.gguf",
            "stepfun-ai_Step-3.5-Flash-Q8_0/stepfun-ai_Step-3.5-Flash-Q8_0-00005-of-00006.gguf",
            "stepfun-ai_Step-3.5-Flash-Q8_0/stepfun-ai_Step-3.5-Flash-Q8_0-00006-of-00006.gguf"
          ]
        }
      ],
      "lab": "StepFun",
      "familyId": "stepfun-step-3-5-flash",
      "familyName": "Step 3.5 Flash",
      "likes": 838,
      "sourceDownloads": 195225,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.060862914019011345,
        "source": "https://huggingface.co/stepfun-ai/Step-3.5-Flash"
      },
      "edition": "step-3.5-flash",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/gpt-oss-20b-GGUF",
      "id": "openai-gpt-oss-20b",
      "name": "gpt-oss-20b",
      "sourceRepo": "openai/gpt-oss-20b",
      "parameters": 20914757184,
      "layers": 24,
      "kvHeads": 8,
      "headDim": 64,
      "context": 131072,
      "contextSource": "https://huggingface.co/openai/gpt-oss-20b/raw/6cee5e81ee83917806bbde320786a8fb61efebee/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 12,
            "bytesPerToken": 2048,
            "window": 128
          },
          {
            "layers": 12,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "apache-2.0",
      "released": "2025-08-04",
      "downloads": 488254,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "d449b42d93e1c2c7bda5312f5c25c8fb91dfa9b4",
      "configUrl": "https://huggingface.co/openai/gpt-oss-20b/raw/6cee5e81ee83917806bbde320786a8fb61efebee/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 11624759488,
          "files": [
            "gpt-oss-20b-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 12109567168,
          "files": [
            "gpt-oss-20b-Q8_0.gguf"
          ]
        }
      ],
      "lab": "OpenAI",
      "familyId": "openai-gpt-oss",
      "familyName": "gpt-oss",
      "likes": 5122,
      "sourceDownloads": 6574377,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.17212726728446248,
        "activeParameters": 3600000000,
        "source": "https://huggingface.co/openai/gpt-oss-20b"
      },
      "edition": "gpt-oss-20b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/gpt-oss-120b-GGUF",
      "id": "openai-gpt-oss-120b",
      "name": "gpt-oss-120b",
      "sourceRepo": "openai/gpt-oss-120b",
      "parameters": 116829156672,
      "layers": 36,
      "kvHeads": 8,
      "headDim": 64,
      "context": 131072,
      "contextSource": "https://huggingface.co/openai/gpt-oss-120b/raw/b5c939de8f754692c1647ca79fbf85e8c1e70f8a/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "windowed",
        "groups": [
          {
            "layers": 18,
            "bytesPerToken": 2048,
            "window": 128
          },
          {
            "layers": 18,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0,
        "note": "Assumes a runner with sliding-window KV caching. Shared-layer cache savings are not deducted."
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "apache-2.0",
      "released": "2025-08-04",
      "downloads": 72545,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "ff1a82da6ad466e32284fa3d2b86694db3204789",
      "configUrl": "https://huggingface.co/openai/gpt-oss-120b/raw/b5c939de8f754692c1647ca79fbf85e8c1e70f8a/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 62768723552,
          "files": [
            "Q4_K_M/gpt-oss-120b-Q4_K_M-00001-of-00002.gguf",
            "Q4_K_M/gpt-oss-120b-Q4_K_M-00002-of-00002.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 63387347520,
          "files": [
            "Q8_0/gpt-oss-120b-Q8_0-00001-of-00002.gguf",
            "Q8_0/gpt-oss-120b-Q8_0-00002-of-00002.gguf"
          ]
        }
      ],
      "lab": "OpenAI",
      "familyId": "openai-gpt-oss",
      "familyName": "gpt-oss",
      "likes": 5346,
      "sourceDownloads": 4481138,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.04365348638369738,
        "activeParameters": 5100000000,
        "source": "https://huggingface.co/openai/gpt-oss-120b"
      },
      "edition": "gpt-oss-120b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "sourceRepo": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
      "repo": "unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF",
      "id": "meta-llama-llama-4-scout-17b-16e-instruct",
      "name": "Llama-4-Scout-17B-16E",
      "configRepo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "parameters": 107769861184,
      "layers": 48,
      "kvHeads": 8,
      "headDim": 128,
      "context": 10485760,
      "contextSource": "https://huggingface.co/unsloth/Llama-4-Scout-17B-16E-Instruct/raw/afd8e498c87bda51c7ea8ec68ea2f7c066e6340b/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 1746780928
      },
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 48,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "other",
      "released": "2025-04-02",
      "downloads": 27771,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "72a6853f56a66dc13a3a4b6bdc9cf7ee4c364b47",
      "configUrl": "https://huggingface.co/unsloth/Llama-4-Scout-17B-16E-Instruct/raw/afd8e498c87bda51c7ea8ec68ea2f7c066e6340b/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 65359900352,
          "files": [
            "Q4_K_M/Llama-4-Scout-17B-16E-Instruct-Q4_K_M-00001-of-00002.gguf",
            "Q4_K_M/Llama-4-Scout-17B-16E-Instruct-Q4_K_M-00002-of-00002.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 114531589472,
          "files": [
            "Q8_0/Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003.gguf",
            "Q8_0/Llama-4-Scout-17B-16E-Instruct-Q8_0-00002-of-00003.gguf",
            "Q8_0/Llama-4-Scout-17B-16E-Instruct-Q8_0-00003-of-00003.gguf"
          ]
        }
      ],
      "lab": "Meta",
      "familyId": "meta-llama-4",
      "familyName": "Llama 4",
      "likes": 1363,
      "sourceDownloads": 177818,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.15774354548880032,
        "activeParameters": 17000000000,
        "source": "https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct"
      },
      "edition": "llama-4-scout-17b-16e",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "sourceRepo": "meta-llama/Llama-4-Maverick-17B-128E-Instruct",
      "repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF",
      "id": "meta-llama-llama-4-maverick-17b-128e-instruct",
      "name": "Llama-4-Maverick-17B-128E",
      "configRepo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "parameters": 400711848960,
      "layers": 48,
      "kvHeads": 8,
      "headDim": 128,
      "context": 1048576,
      "contextSource": "https://huggingface.co/unsloth/Llama-4-Maverick-17B-128E-Instruct/raw/86c5ecb6f2fbddf604c60120125fb202e09f2556/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 1746780960
      },
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 48,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "other",
      "released": "2025-04-01",
      "downloads": 9229,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "41032e5471dd6ea5b349062d978626215d6c5bba",
      "configUrl": "https://huggingface.co/unsloth/Llama-4-Maverick-17B-128E-Instruct/raw/86c5ecb6f2fbddf604c60120125fb202e09f2556/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 242767153440,
          "files": [
            "Q4_K_M/Llama-4-Maverick-17B-128E-Instruct-Q4_K_M-00001-of-00005.gguf",
            "Q4_K_M/Llama-4-Maverick-17B-128E-Instruct-Q4_K_M-00002-of-00005.gguf",
            "Q4_K_M/Llama-4-Maverick-17B-128E-Instruct-Q4_K_M-00003-of-00005.gguf",
            "Q4_K_M/Llama-4-Maverick-17B-128E-Instruct-Q4_K_M-00004-of-00005.gguf",
            "Q4_K_M/Llama-4-Maverick-17B-128E-Instruct-Q4_K_M-00005-of-00005.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 425817094048,
          "files": [
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00001-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00002-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00003-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00004-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00005-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00006-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00007-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00008-of-00009.gguf",
            "Q8_0/Llama-4-Maverick-17B-128E-Instruct-Q8_0-00009-of-00009.gguf"
          ]
        }
      ],
      "lab": "Meta",
      "familyId": "meta-llama-4",
      "familyName": "Llama 4",
      "likes": 514,
      "sourceDownloads": 9720,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.042424500408763756,
        "activeParameters": 17000000000,
        "source": "https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct"
      },
      "edition": "llama-4-maverick-17b-128e",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "id": "meta-llama-llama-3-3-70b-instruct",
      "name": "Llama-3.3-70B",
      "sourceRepo": "meta-llama/Llama-3.3-70B-Instruct",
      "configRepo": "unsloth/Llama-3.3-70B-Instruct",
      "repo": "bartowski/Llama-3.3-70B-Instruct-GGUF",
      "tasks": [
        "chat"
      ],
      "note": "An instruction-tuned assistant for conversation, writing, and summarization. Subject to the Llama community license.",
      "parameters": 70553706560,
      "layers": 80,
      "kvHeads": 8,
      "headDim": 128,
      "context": 131072,
      "license": "llama3.3",
      "revision": "b6c5c9f176f3279204034e1d16d393105e95cb88",
      "configUrl": "https://huggingface.co/unsloth/Llama-3.3-70B-Instruct/raw/99cd0d2c829e92a67c844f9144c2509632e5c87f/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 42520398816,
          "files": [
            "Llama-3.3-70B-Instruct-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 74975055008,
          "files": [
            "Llama-3.3-70B-Instruct-Q8_0/Llama-3.3-70B-Instruct-Q8_0-00001-of-00002.gguf",
            "Llama-3.3-70B-Instruct-Q8_0/Llama-3.3-70B-Instruct-Q8_0-00002-of-00002.gguf"
          ]
        }
      ],
      "contextSource": "https://huggingface.co/unsloth/Llama-3.3-70B-Instruct/raw/99cd0d2c829e92a67c844f9144c2509632e5c87f/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 80,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "released": "2024-11-26",
      "downloads": 17431,
      "lab": "Meta",
      "familyId": "meta-llama-3-3",
      "familyName": "Llama 3.3",
      "likes": 3092,
      "sourceDownloads": 414491,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "llama-3.3-70b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "id": "meta-llama-llama-3-2-3b-instruct",
      "name": "Llama-3.2-3B",
      "sourceRepo": "meta-llama/Llama-3.2-3B-Instruct",
      "configRepo": "unsloth/Llama-3.2-3B-Instruct",
      "repo": "bartowski/Llama-3.2-3B-Instruct-GGUF",
      "tasks": [
        "chat"
      ],
      "note": "An instruction-tuned assistant for conversation, writing, and summarization. Subject to the Llama community license.",
      "parameters": 3212749888,
      "layers": 28,
      "kvHeads": 8,
      "headDim": 128,
      "context": 131072,
      "license": "llama3.2",
      "revision": "5ab33fa94d1d04e903623ae72c95d1696f09f9e8",
      "configUrl": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/raw/006f5dcd1393c3add266de40994ba96225e9689d/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 2019377696,
          "files": [
            "Llama-3.2-3B-Instruct-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 3421899296,
          "files": [
            "Llama-3.2-3B-Instruct-Q8_0.gguf"
          ]
        }
      ],
      "contextSource": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/raw/006f5dcd1393c3add266de40994ba96225e9689d/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 28,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "released": "2024-09-18",
      "downloads": 143954,
      "lab": "Meta",
      "familyId": "meta-llama-3-2",
      "familyName": "Llama 3.2",
      "likes": 2722,
      "sourceDownloads": 1611319,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "llama-3.2-3b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "id": "meta-llama-llama-3-2-1b-instruct",
      "name": "Llama-3.2-1B",
      "sourceRepo": "meta-llama/Llama-3.2-1B-Instruct",
      "configRepo": "unsloth/Llama-3.2-1B-Instruct",
      "repo": "bartowski/Llama-3.2-1B-Instruct-GGUF",
      "tasks": [
        "chat"
      ],
      "note": "An instruction-tuned assistant for conversation, writing, and summarization. Subject to the Llama community license.",
      "parameters": 1235814432,
      "layers": 16,
      "kvHeads": 8,
      "headDim": 64,
      "context": 131072,
      "license": "llama3.2",
      "revision": "067b946cf014b7c697f3654f621d577a3e3afd1c",
      "configUrl": "https://huggingface.co/unsloth/Llama-3.2-1B-Instruct/raw/5a8abab4a5d6f164389b1079fb721cfab8d7126c/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 807694464,
          "files": [
            "Llama-3.2-1B-Instruct-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 1321083008,
          "files": [
            "Llama-3.2-1B-Instruct-Q8_0.gguf"
          ]
        }
      ],
      "contextSource": "https://huggingface.co/unsloth/Llama-3.2-1B-Instruct/raw/5a8abab4a5d6f164389b1079fb721cfab8d7126c/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 16,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0
      },
      "released": "2024-09-18",
      "downloads": 113025,
      "lab": "Meta",
      "familyId": "meta-llama-3-2",
      "familyName": "Llama 3.2",
      "likes": 1770,
      "sourceDownloads": 7743934,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "llama-3.2-1b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/Fara1.5-27B-GGUF",
      "id": "microsoft-fara1-5-27b",
      "name": "Fara1.5-27B",
      "sourceRepo": "microsoft/Fara1.5-27B",
      "parameters": 26895998464,
      "layers": 64,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/microsoft/Fara1.5-27B/raw/299c8406a6c6256d45ec200d1ac12b34c5599d9b/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-Fara1.5-27B-f16.gguf",
        "bytes": 927607456
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 16,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 158859264
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "mit",
      "released": "2026-07-17",
      "downloads": 87089,
      "note": "A model for computer use and screen understanding. Select Vision to include its image encoder.",
      "revision": "dd7cba968d1a9c8feab0c2b85d93b117e6cc16fe",
      "configUrl": "https://huggingface.co/microsoft/Fara1.5-27B/raw/299c8406a6c6256d45ec200d1ac12b34c5599d9b/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 17533552448,
          "files": [
            "Fara1.5-27B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 28665067328,
          "files": [
            "Fara1.5-27B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Microsoft",
      "familyId": "microsoft-fara-1-5",
      "familyName": "Fara 1.5",
      "likes": 292,
      "sourceDownloads": 1966,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "fara1.5-27b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/Fara1.5-4B-GGUF",
      "id": "microsoft-fara1-5-4b",
      "name": "Fara1.5-4B",
      "sourceRepo": "microsoft/Fara1.5-4B",
      "parameters": 4205751296,
      "layers": 32,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/microsoft/Fara1.5-4B/raw/776a33ae5b2ad503796a97ae20fdc66f61d2feea/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-Fara1.5-4B-f16.gguf",
        "bytes": 672423584
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 8,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 53477376
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "mit",
      "released": "2026-07-17",
      "downloads": 85571,
      "note": "A model for computer use and screen understanding. Select Vision to include its image encoder.",
      "revision": "b97f335231e01efbbad37bb89b5310340fc10735",
      "configUrl": "https://huggingface.co/microsoft/Fara1.5-4B/raw/776a33ae5b2ad503796a97ae20fdc66f61d2feea/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 2884850784,
          "files": [
            "Fara1.5-4B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 4493954144,
          "files": [
            "Fara1.5-4B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Microsoft",
      "familyId": "microsoft-fara-1-5",
      "familyName": "Fara 1.5",
      "likes": 47,
      "sourceDownloads": 1650,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "fara1.5-4b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/Fara1.5-9B-GGUF",
      "id": "microsoft-fara1-5-9b",
      "name": "Fara1.5-9B",
      "sourceRepo": "microsoft/Fara1.5-9B",
      "parameters": 8953803264,
      "layers": 32,
      "kvHeads": 4,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/microsoft/Fara1.5-9B/raw/1a93677cd89d5601bc2ed759791e981f3a520032/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-Fara1.5-9B-f16.gguf",
        "bytes": 918166048
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 8,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 53477376
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "mit",
      "released": "2026-05-12",
      "downloads": 86797,
      "note": "A model for computer use and screen understanding. Select Vision to include its image encoder.",
      "revision": "153cb27ac91d4a2b9391ecf278542e610d040178",
      "configUrl": "https://huggingface.co/microsoft/Fara1.5-9B/raw/1a93677cd89d5601bc2ed759791e981f3a520032/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 5910783104,
          "files": [
            "Fara1.5-9B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 9545983104,
          "files": [
            "Fara1.5-9B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Microsoft",
      "familyId": "microsoft-fara-1-5",
      "familyName": "Fara 1.5",
      "likes": 47,
      "sourceDownloads": 4579,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "fara1.5-9b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "sourceRepo": "microsoft/Phi-4-reasoning-plus",
      "repo": "unsloth/Phi-4-reasoning-plus-GGUF",
      "id": "microsoft-phi-4-reasoning-plus",
      "name": "Phi-4-reasoning-plus",
      "parameters": 14659507200,
      "layers": 40,
      "kvHeads": 10,
      "headDim": 128,
      "context": 32768,
      "contextSource": "https://huggingface.co/microsoft/Phi-4-reasoning-plus/raw/69baf8528e1bcf05f475034d9e5dd32875ed125f/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 40,
            "bytesPerToken": 5120
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding",
        "reasoning"
      ],
      "license": "mit",
      "released": "2025-04-17",
      "downloads": 7783,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "80fff8542dc7b88dba725b660beefd80e91e80c9",
      "configUrl": "https://huggingface.co/microsoft/Phi-4-reasoning-plus/raw/69baf8528e1bcf05f475034d9e5dd32875ed125f/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 9053117120,
          "files": [
            "Phi-4-reasoning-plus-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 15580502720,
          "files": [
            "Phi-4-reasoning-plus-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Microsoft",
      "familyId": "microsoft-phi-4-reasoning",
      "familyName": "Phi 4 Reasoning",
      "likes": 348,
      "sourceDownloads": 9991,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "phi-4-reasoning-plus",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/granite-4.2-30b-GGUF",
      "id": "ibm-granite-granite-4-2-30b",
      "name": "granite-4.2-30b",
      "sourceRepo": "ibm-granite/granite-4.2-30b",
      "parameters": 29276770304,
      "layers": 64,
      "kvHeads": 8,
      "headDim": 128,
      "context": 131072,
      "contextSource": "https://huggingface.co/ibm-granite/granite-4.2-30b/raw/9e668ce1c538387ef24d3644e9b0606647762636/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 64,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-08-07",
      "downloads": 3525,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "1847d3b70241af9d656f382a4cf29d5c6573e584",
      "configUrl": "https://huggingface.co/ibm-granite/granite-4.2-30b/raw/9e668ce1c538387ef24d3644e9b0606647762636/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 18027639840,
          "files": [
            "granite-4.2-30b-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 31111705632,
          "files": [
            "granite-4.2-30b-Q8_0.gguf"
          ]
        }
      ],
      "lab": "IBM",
      "familyId": "ibm-granite-4-2",
      "familyName": "Granite 4.2",
      "likes": 127,
      "sourceDownloads": 35751,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "granite-4.2-30b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/granite-4.2-8b-GGUF",
      "id": "ibm-granite-granite-4-2-8b",
      "name": "granite-4.2-8b",
      "sourceRepo": "ibm-granite/granite-4.2-8b",
      "parameters": 8791592960,
      "layers": 40,
      "kvHeads": 8,
      "headDim": 128,
      "context": 131072,
      "contextSource": "https://huggingface.co/ibm-granite/granite-4.2-8b/raw/f8de16cdcdbc6c779ca517604e050d82cc119e44/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 40,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-08-07",
      "downloads": 4353,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "a592100df8fe4931c7cffbac7b28e8176a1d52da",
      "configUrl": "https://huggingface.co/ibm-granite/granite-4.2-8b/raw/f8de16cdcdbc6c779ca517604e050d82cc119e44/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 5539283360,
          "files": [
            "granite-4.2-8b-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 9345614240,
          "files": [
            "granite-4.2-8b-Q8_0.gguf"
          ]
        }
      ],
      "lab": "IBM",
      "familyId": "ibm-granite-4-2",
      "familyName": "Granite 4.2",
      "likes": 93,
      "sourceDownloads": 128378,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "granite-4.2-8b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/granite-4.2-3b-GGUF",
      "id": "ibm-granite-granite-4-2-3b",
      "name": "granite-4.2-3b",
      "sourceRepo": "ibm-granite/granite-4.2-3b",
      "parameters": 3659737600,
      "layers": 40,
      "kvHeads": 8,
      "headDim": 64,
      "context": 131072,
      "contextSource": "https://huggingface.co/ibm-granite/granite-4.2-3b/raw/e459acceac81e5fe67c07d9cfc72329a332e7eb1/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 40,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-08-07",
      "downloads": 3864,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "4093456941a783ab5d8268e00a7725532c1e9af3",
      "configUrl": "https://huggingface.co/ibm-granite/granite-4.2-3b/raw/e459acceac81e5fe67c07d9cfc72329a332e7eb1/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 2317126048,
          "files": [
            "granite-4.2-3b-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 3892652448,
          "files": [
            "granite-4.2-3b-Q8_0.gguf"
          ]
        }
      ],
      "lab": "IBM",
      "familyId": "ibm-granite-4-2",
      "familyName": "Granite 4.2",
      "likes": 106,
      "sourceDownloads": 50610,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "granite-4.2-3b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/granite-4.1-30b-GGUF",
      "id": "ibm-granite-granite-4-1-30b",
      "name": "granite-4.1-30b",
      "sourceRepo": "ibm-granite/granite-4.1-30b",
      "parameters": 28865728512,
      "layers": 64,
      "kvHeads": 8,
      "headDim": 128,
      "context": 131072,
      "contextSource": "https://huggingface.co/ibm-granite/granite-4.1-30b/raw/4fae6278f7132abf5e971f9de49ebbad09c54cce/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 64,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-04-06",
      "downloads": 1922,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "6cb34f31b11ca4c1433de1af7391dac46de4e666",
      "configUrl": "https://huggingface.co/ibm-granite/granite-4.1-30b/raw/4fae6278f7132abf5e971f9de49ebbad09c54cce/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 17490241472,
          "files": [
            "granite-4.1-30b-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 30674970560,
          "files": [
            "granite-4.1-30b-Q8_0.gguf"
          ]
        }
      ],
      "lab": "IBM",
      "familyId": "ibm-granite-4-1",
      "familyName": "Granite 4.1",
      "likes": 147,
      "sourceDownloads": 301518,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "granite-4.1-30b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/granite-4.1-8b-GGUF",
      "id": "ibm-granite-granite-4-1-8b",
      "name": "granite-4.1-8b",
      "sourceRepo": "ibm-granite/granite-4.1-8b",
      "parameters": 8791592960,
      "layers": 40,
      "kvHeads": 8,
      "headDim": 128,
      "context": 131072,
      "contextSource": "https://huggingface.co/ibm-granite/granite-4.1-8b/raw/1504002f650e656a0a3789d99574df12e3e94ed0/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 40,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-04-06",
      "downloads": 4695,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "6f9671f73eb03273bc09319194b8a4e810e03a8f",
      "configUrl": "https://huggingface.co/ibm-granite/granite-4.1-8b/raw/1504002f650e656a0a3789d99574df12e3e94ed0/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 5347915136,
          "files": [
            "granite-4.1-8b-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 9345611136,
          "files": [
            "granite-4.1-8b-Q8_0.gguf"
          ]
        }
      ],
      "lab": "IBM",
      "familyId": "ibm-granite-4-1",
      "familyName": "Granite 4.1",
      "likes": 256,
      "sourceDownloads": 179276,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "granite-4.1-8b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/granite-4.1-3b-GGUF",
      "id": "ibm-granite-granite-4-1-3b",
      "name": "granite-4.1-3b",
      "sourceRepo": "ibm-granite/granite-4.1-3b",
      "parameters": 3402836480,
      "layers": 40,
      "kvHeads": 8,
      "headDim": 64,
      "context": 131072,
      "contextSource": "https://huggingface.co/ibm-granite/granite-4.1-3b/raw/c0650403e44e78ec0262dab1c90914c65b196c4e/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 40,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-04-06",
      "downloads": 5170,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "5b88826e4b80789548180f8faab39c5cf68772c9",
      "configUrl": "https://huggingface.co/ibm-granite/granite-4.1-3b/raw/c0650403e44e78ec0262dab1c90914c65b196c4e/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 2099502400,
          "files": [
            "granite-4.1-3b-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 3619691840,
          "files": [
            "granite-4.1-3b-Q8_0.gguf"
          ]
        }
      ],
      "lab": "IBM",
      "familyId": "ibm-granite-4-1",
      "familyName": "Granite 4.1",
      "likes": 112,
      "sourceDownloads": 521815,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "granite-4.1-3b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/LFM2.5-VL-3B-GGUF",
      "id": "liquidai-lfm2-5-vl-3b",
      "name": "LFM2.5-VL-3B",
      "sourceRepo": "LiquidAI/LFM2.5-VL-3B",
      "parameters": 2697198592,
      "layers": 30,
      "kvHeads": 8,
      "headDim": 64,
      "context": 32768,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-VL-3B/raw/a3af5799199acdd2a4f56ac4342816abb46c12a9/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 853994080
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 8,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 540672
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "other",
      "released": "2026-08-11",
      "downloads": 19760,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "22f063714556bd4daa81c40dc1ff15ca37ae92ea",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-VL-3B/raw/35a118d938ce6d123ac2d371649f24a8efb69058/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 1674455424,
          "files": [
            "LFM2.5-VL-3B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 2874780032,
          "files": [
            "LFM2.5-VL-3B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5-vision",
      "familyName": "LFM 2.5 Vision",
      "likes": 214,
      "sourceDownloads": 26745,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "lfm2.5-vl-3b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/LiquidAI_LFM2.5-2.6B-GGUF",
      "id": "liquidai-lfm2-5-2-6b",
      "name": "LFM2.5-2.6B",
      "sourceRepo": "LiquidAI/LFM2.5-2.6B",
      "parameters": 2697198592,
      "layers": 30,
      "kvHeads": 8,
      "headDim": 64,
      "context": 131072,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-2.6B/raw/654f9463ce32b05d0429d76fe1f580b27d4c1ac0/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 8,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 540672
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "other",
      "released": "2026-07-28",
      "downloads": 3495,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "0446ca3a34aea723479864cde5fdb9ef0a8fb3e8",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-2.6B/raw/654f9463ce32b05d0429d76fe1f580b27d4c1ac0/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 1684023584,
          "files": [
            "LiquidAI_LFM2.5-2.6B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 2874779936,
          "files": [
            "LiquidAI_LFM2.5-2.6B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5",
      "familyName": "LFM 2.5",
      "likes": 794,
      "sourceDownloads": 108072,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "lfm2.5-2.6b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/LFM2.5-230M-GGUF",
      "id": "liquidai-lfm2-5-230m",
      "name": "LFM2.5-230M",
      "sourceRepo": "LiquidAI/LFM2.5-230M",
      "parameters": 229693184,
      "layers": 14,
      "kvHeads": 8,
      "headDim": 64,
      "context": 128000,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-230M/raw/40cb2ad3b3044d5a41eee083a6103c8b523afa45/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 98304
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "other",
      "released": "2026-06-24",
      "downloads": 3103,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "c68127fd28200f50f79e41738bc7f1e7888862ec",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-230M/raw/40cb2ad3b3044d5a41eee083a6103c8b523afa45/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 153406656,
          "files": [
            "LFM2.5-230M-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 246598848,
          "files": [
            "LFM2.5-230M-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5",
      "familyName": "LFM 2.5",
      "likes": 301,
      "sourceDownloads": 88558,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "lfm2.5-230m",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/LFM2.5-8B-A1B-GGUF",
      "id": "liquidai-lfm2-5-8b-a1b",
      "name": "LFM2.5-8B-A1B",
      "sourceRepo": "LiquidAI/LFM2.5-8B-A1B",
      "parameters": 8467856832,
      "layers": 24,
      "kvHeads": 8,
      "headDim": 64,
      "context": 128000,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-8B-A1B/raw/5dd22602c2e9f6a097b1de4c4efe0658b605015c/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 442368
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "other",
      "released": "2026-05-28",
      "downloads": 20110,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "37563684c4a625e9958906fd924e15095cf9c896",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-8B-A1B/raw/5dd22602c2e9f6a097b1de4c4efe0658b605015c/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "UD-Q4_K_M",
          "precision": 4,
          "bytes": 5322223200,
          "files": [
            "LFM2.5-8B-A1B-UD-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 9010196064,
          "files": [
            "LFM2.5-8B-A1B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5",
      "familyName": "LFM 2.5",
      "likes": 779,
      "sourceDownloads": 32425,
      "throughput": {
        "kind": "moe",
        "weightFraction": 0.11809363571441166,
        "activeParameters": 1000000000,
        "source": "https://huggingface.co/LiquidAI/LFM2.5-8B-A1B"
      },
      "edition": "lfm2.5-8b-a1b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "LiquidAI/LFM2.5-VL-450M-GGUF",
      "id": "liquidai-lfm2-5-vl-450m",
      "name": "LFM2.5-VL-450M",
      "sourceRepo": "LiquidAI/LFM2.5-VL-450M",
      "parameters": 354483968,
      "layers": 16,
      "kvHeads": 8,
      "headDim": 64,
      "context": 128000,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-VL-450M/raw/fc6221ca597f3315e4f82fc2df606783267b34ba/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-LFM2.5-VL-450m-F16.gguf",
        "bytes": 189126080
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 122880
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "other",
      "released": "2026-04-08",
      "downloads": 19149,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "1abed04b6fe71314d8c446a1371c03d7c332266d",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-VL-450M/raw/fc6221ca597f3315e4f82fc2df606783267b34ba/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 229313568,
          "files": [
            "LFM2.5-VL-450M-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 379219104,
          "files": [
            "LFM2.5-VL-450M-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5-vision",
      "familyName": "LFM 2.5 Vision",
      "likes": 225,
      "sourceDownloads": 35526,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "lfm2.5-vl-450m",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "LiquidAI/LFM2.5-350M-GGUF",
      "id": "liquidai-lfm2-5-350m",
      "name": "LFM2.5-350M",
      "sourceRepo": "LiquidAI/LFM2.5-350M",
      "parameters": 354483968,
      "layers": 16,
      "kvHeads": 8,
      "headDim": 64,
      "context": 128000,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-350M/raw/9e6c6ccf47cd318696e137d381a7ded8fe4df09f/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 122880
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "other",
      "released": "2026-03-31",
      "downloads": 74772,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "657e078c94084481950a2d555a941481f715536b",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-350M/raw/9e6c6ccf47cd318696e137d381a7ded8fe4df09f/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 229312224,
          "files": [
            "LFM2.5-350M-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 379217632,
          "files": [
            "LFM2.5-350M-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5",
      "familyName": "LFM 2.5",
      "likes": 430,
      "sourceDownloads": 69943,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "lfm2.5-350m",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/LFM2.5-1.2B-Instruct-GGUF",
      "id": "liquidai-lfm2-5-1-2b-instruct",
      "name": "LFM2.5-1.2B",
      "sourceRepo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "parameters": 1170340608,
      "layers": 16,
      "kvHeads": 8,
      "headDim": 64,
      "context": 128000,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct/raw/0f604ada3f766f9f257460c4c9f0b5d6f69d431b/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 245760
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "other",
      "released": "2026-01-06",
      "downloads": 36254,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "b01cc872e5918b0fdec170003cd854851e3ed854",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct/raw/0f604ada3f766f9f257460c4c9f0b5d6f69d431b/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 730895584,
          "files": [
            "LFM2.5-1.2B-Instruct-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 1246254304,
          "files": [
            "LFM2.5-1.2B-Instruct-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5",
      "familyName": "LFM 2.5",
      "likes": 673,
      "sourceDownloads": 119209,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "lfm2.5-1.2b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "unsloth/LFM2.5-VL-1.6B-GGUF",
      "id": "liquidai-lfm2-5-vl-1-6b",
      "name": "LFM2.5-VL-1.6B",
      "sourceRepo": "LiquidAI/LFM2.5-VL-1.6B",
      "parameters": 1170340608,
      "layers": 16,
      "kvHeads": 8,
      "headDim": 64,
      "context": 128000,
      "contextSource": "https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B/raw/919fde3d022e3f90a4716006f993938ee8c2eb97/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-F16.gguf",
        "bytes": 853993984
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 245760
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "other",
      "released": "2026-01-05",
      "downloads": 3045,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "84de8c11b63ef084383e6dc1e565843b58c5c783",
      "configUrl": "https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B/raw/919fde3d022e3f90a4716006f993938ee8c2eb97/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 730895968,
          "files": [
            "LFM2.5-VL-1.6B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 1246254688,
          "files": [
            "LFM2.5-VL-1.6B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "Liquid AI",
      "familyId": "liquid-ai-lfm-2-5-vision",
      "familyName": "LFM 2.5 Vision",
      "likes": 329,
      "sourceDownloads": 28333,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "lfm2.5-vl-1.6b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "bartowski/MiniCPM5-2B-GGUF",
      "id": "openbmb-minicpm5-2b",
      "name": "MiniCPM5-2B",
      "sourceRepo": "openbmb/MiniCPM5-2B",
      "parameters": 2516756480,
      "layers": 42,
      "kvHeads": 2,
      "headDim": 128,
      "context": 131072,
      "contextSource": "https://huggingface.co/openbmb/MiniCPM5-2B/raw/12a3808a956f869c767195e9266b59c4d21d92e2/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 42,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-09-06",
      "downloads": 24477,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "31fee0334c4ed9f1e079afe586ba30cb83e0f883",
      "configUrl": "https://huggingface.co/openbmb/MiniCPM5-2B/raw/f97400052a43d642bbc6e9975e2397e3ae6a6b52/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 1615826144,
          "files": [
            "MiniCPM5-2B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 2679712992,
          "files": [
            "MiniCPM5-2B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "OpenBMB",
      "familyId": "openbmb-minicpm-5",
      "familyName": "MiniCPM 5",
      "likes": 1712,
      "sourceDownloads": 1059242,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "minicpm5-2b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "openbmb/MiniCPM5-1B-GGUF",
      "id": "openbmb-minicpm5-1b",
      "name": "MiniCPM5-1B",
      "sourceRepo": "openbmb/MiniCPM5-1B",
      "parameters": 1080632832,
      "layers": 24,
      "kvHeads": 2,
      "headDim": 128,
      "context": 131072,
      "contextSource": "https://huggingface.co/openbmb/MiniCPM5-1B/raw/87179e5c1f455ef22e6223592d2d61351b525bfc/config.json",
      "vision": false,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 24,
            "bytesPerToken": 1024
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-05-21",
      "downloads": 38210,
      "note": "An open-weight language model. See the publisher’s model card for its capabilities, usage instructions, and license.",
      "revision": "3d55fac80935ae6456986ad2384b5cbcc4d6c948",
      "configUrl": "https://huggingface.co/openbmb/MiniCPM5-1B/raw/87179e5c1f455ef22e6223592d2d61351b525bfc/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 688065920,
          "files": [
            "MiniCPM5-1B-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 1153529216,
          "files": [
            "MiniCPM5-1B-Q8_0.gguf"
          ]
        }
      ],
      "lab": "OpenBMB",
      "familyId": "openbmb-minicpm-5",
      "familyName": "MiniCPM 5",
      "likes": 1148,
      "sourceDownloads": 436837,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "minicpm5-1b",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "openbmb/MiniCPM-V-4.6-gguf",
      "id": "openbmb-minicpm-v-4-6",
      "name": "MiniCPM-V-4.6",
      "sourceRepo": "openbmb/MiniCPM-V-4.6",
      "parameters": 752161600,
      "layers": 24,
      "kvHeads": 2,
      "headDim": 256,
      "context": 262144,
      "contextSource": "https://huggingface.co/openbmb/MiniCPM-V-4.6/raw/36f34a661a4bd35d0dc2294cb044d2584646c7d3/config.json",
      "vision": true,
      "projector": {
        "file": "mmproj-model-f16.gguf",
        "bytes": 1108746944
      },
      "memory": {
        "kind": "hybrid",
        "groups": [
          {
            "layers": 6,
            "bytesPerToken": 2048
          }
        ],
        "stateBytes": 20643840
      },
      "tasks": [
        "chat",
        "coding",
        "vision"
      ],
      "license": "apache-2.0",
      "released": "2026-04-13",
      "downloads": 21660,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "afe9accb78d2995d214cd912920c9c92f4015faa",
      "configUrl": "https://huggingface.co/openbmb/MiniCPM-V-4.6/raw/36f34a661a4bd35d0dc2294cb044d2584646c7d3/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 529101504,
          "files": [
            "MiniCPM-V-4_6-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 811591616,
          "files": [
            "MiniCPM-V-4_6-Q8_0.gguf"
          ]
        }
      ],
      "lab": "OpenBMB",
      "familyId": "openbmb-minicpm-v-4-6",
      "familyName": "MiniCPM V 4.6",
      "likes": 1227,
      "sourceDownloads": 286988,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "minicpm-v-4.6",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    },
    {
      "repo": "openbmb/MiniCPM-o-4_5-gguf",
      "id": "openbmb-minicpm-o-4-5",
      "name": "MiniCPM-o-4_5",
      "sourceRepo": "openbmb/MiniCPM-o-4_5",
      "parameters": 8189195264,
      "layers": 36,
      "kvHeads": 8,
      "headDim": 128,
      "context": 40960,
      "contextSource": "https://huggingface.co/openbmb/MiniCPM-o-4_5/raw/503e754207c94da6bb26850b4469f367c9ea3582/config.json",
      "vision": true,
      "projector": null,
      "memory": {
        "kind": "attention",
        "groups": [
          {
            "layers": 36,
            "bytesPerToken": 4096
          }
        ],
        "stateBytes": 0
      },
      "tasks": [
        "chat",
        "coding"
      ],
      "license": "apache-2.0",
      "released": "2026-02-03",
      "downloads": 51507,
      "note": "A model for text and image understanding. Select Vision to include its image encoder in the memory estimate.",
      "revision": "db25077c33951fe163b42986fba0132e279872a2",
      "configUrl": "https://huggingface.co/openbmb/MiniCPM-o-4_5/raw/503e754207c94da6bb26850b4469f367c9ea3582/config.json",
      "verifiedAt": "2026-10-04",
      "variants": [
        {
          "quant": "Q4_K_M",
          "precision": 4,
          "bytes": 5026714400,
          "files": [
            "MiniCPM-o-4_5-Q4_K_M.gguf"
          ]
        },
        {
          "quant": "Q8_0",
          "precision": 8,
          "bytes": 8707877536,
          "files": [
            "MiniCPM-o-4_5-Q8_0.gguf"
          ]
        }
      ],
      "lab": "OpenBMB",
      "familyId": "openbmb-minicpm-o-4-5",
      "familyName": "MiniCPM o 4.5",
      "likes": 1503,
      "sourceDownloads": 735210,
      "throughput": {
        "kind": "dense",
        "weightFraction": 1
      },
      "edition": "minicpm-o-4_5",
      "expiresAt": "2026-10-11T09:45:02.617Z"
    }
  ],
  "popularityUpdatedAt": "2026-10-04T09:45:02.617Z",
  "releasesCheckedAt": "2026-10-04T09:45:02.617Z",
  "pending": [],
  "retired": []
}
