{
  "self": "https://aigearwatch.com/models.json",
  "documentation": "https://aigearwatch.com/methodology/",
  "related": {
    "hardware": "https://aigearwatch.com/hardware.json",
    "csv": "https://aigearwatch.com/models.csv",
    "summary": "https://aigearwatch.com/llms.txt"
  },
  "updatedAt": "2026-10-01T14:13:33.577Z",
  "license": "Parameter counts traceable to the linked model cards. Attribution appreciated.",
  "requirementBasis": "Weights at the given bytes-per-weight plus a 20 percent allowance for the KV cache, activations, and the runtime. Long contexts and large batches need more.",
  "models": [
    {
      "id": "qwen3-5-0-8b",
      "name": "Qwen3.5-0.8B",
      "family": "Qwen3.5",
      "paramsB": 0.8,
      "activeParamsB": null,
      "note": "Model card lists 0.8B parameters and a native 262,144-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-5-0-8b",
      "requiresGb": {
        "q4": 0.5,
        "q5": 0.6,
        "q8": 1,
        "fp16": 1.9
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.5-0.8B",
          "title": "Qwen3.5-0.8B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "lfm2-5-1-2b-thinking",
      "name": "LFM2.5-1.2B-Thinking",
      "family": "LFM2.5",
      "paramsB": 1.17,
      "activeParamsB": null,
      "note": "Model card lists 1.17B parameters and a 32,768-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#lfm2-5-1-2b-thinking",
      "requiresGb": {
        "q4": 0.7,
        "q5": 0.9,
        "q8": 1.4,
        "fp16": 2.8
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/LiquidAI/LFM2.5-1.2B-Thinking",
          "title": "LFM2.5-1.2B-Thinking model card",
          "publisher": "Liquid AI / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "qwen3-5-2b",
      "name": "Qwen3.5-2B",
      "family": "Qwen3.5",
      "paramsB": 2,
      "activeParamsB": null,
      "note": "Model card lists 2B parameters and a native 262,144-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-5-2b",
      "requiresGb": {
        "q4": 1.2,
        "q5": 1.5,
        "q8": 2.4,
        "fp16": 4.8
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.5-2B",
          "title": "Qwen3.5-2B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "smollm3-3b",
      "name": "SmolLM3-3B",
      "family": "SmolLM3",
      "paramsB": 3,
      "activeParamsB": null,
      "note": "Model card calls it a 3B parameter model; the current config supports 65,536 tokens and the card documents YaRN extensions for 128K and 256K.",
      "url": "https://aigearwatch.com/can-it-run/#smollm3-3b",
      "requiresGb": {
        "q4": 1.8,
        "q5": 2.3,
        "q8": 3.6,
        "fp16": 7.2
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/HuggingFaceTB/SmolLM3-3B",
          "title": "SmolLM3-3B model card",
          "publisher": "Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "qwen3-5-4b",
      "name": "Qwen3.5-4B",
      "family": "Qwen3.5",
      "paramsB": 4,
      "activeParamsB": null,
      "note": "Model card lists 4B parameters and a native 262,144-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-5-4b",
      "requiresGb": {
        "q4": 2.4,
        "q5": 3,
        "q8": 4.8,
        "fp16": 9.6
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.5-4B",
          "title": "Qwen3.5-4B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "gemma-4-e4b",
      "name": "Gemma 4 E4B",
      "family": "Gemma 4",
      "paramsB": 8,
      "activeParamsB": null,
      "note": "Google lists 4.5B effective parameters and 8B total with embeddings. Sizing uses all 8B resident weights, not the effective count.",
      "url": "https://aigearwatch.com/can-it-run/#gemma-4-e4b",
      "requiresGb": {
        "q4": 4.8,
        "q5": 6,
        "q8": 9.6,
        "fp16": 19.2
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/google/gemma-4-E4B-it",
          "title": "Gemma 4 E4B IT model card",
          "publisher": "Google / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "qwen3-8b",
      "name": "Qwen3-8B",
      "family": "Qwen3",
      "paramsB": 8.2,
      "activeParamsB": null,
      "note": "Model card: \"Number of Parameters: 8.2B\". Context 32,768 natively, 131,072 with YaRN.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-8b",
      "requiresGb": {
        "q4": 4.9,
        "q5": 6.1,
        "q8": 9.8,
        "fp16": 19.7
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3-8B",
          "title": "Qwen3-8B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "qwen3-5-9b",
      "name": "Qwen3.5-9B",
      "family": "Qwen3.5",
      "paramsB": 9,
      "activeParamsB": null,
      "note": "Model card lists 9B parameters and a native 262,144-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-5-9b",
      "requiresGb": {
        "q4": 5.4,
        "q5": 6.8,
        "q8": 10.8,
        "fp16": 21.6
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.5-9B",
          "title": "Qwen3.5-9B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "gemma-4-12b",
      "name": "Gemma 4 12B",
      "family": "Gemma 4",
      "paramsB": 11.95,
      "activeParamsB": null,
      "note": "Google lists 11.95B total parameters and a 256K context window for the encoder-free multimodal model.",
      "url": "https://aigearwatch.com/can-it-run/#gemma-4-12b",
      "requiresGb": {
        "q4": 7.2,
        "q5": 9,
        "q8": 14.3,
        "fp16": 28.7
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/google/gemma-4-12B-it",
          "title": "Gemma 4 12B IT model card",
          "publisher": "Google / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "gpt-oss-20b",
      "name": "gpt-oss-20b",
      "family": "gpt-oss",
      "paramsB": 21,
      "activeParamsB": 3.6,
      "note": "OpenAI: \"21B parameters with 3.6B active parameters\"; card says it can run within 16GB of memory.",
      "url": "https://aigearwatch.com/can-it-run/#gpt-oss-20b",
      "requiresGb": {
        "q4": 12.6,
        "q5": 15.8,
        "q8": 25.2,
        "fp16": 50.4
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/openai/gpt-oss-120b",
          "title": "gpt-oss-120b model card",
          "publisher": "OpenAI / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "mistral-small-3-2-24b",
      "name": "Mistral Small 3.2 24B",
      "family": "Mistral",
      "paramsB": 24,
      "activeParamsB": null,
      "note": "Model card lists 24B params. Mistral does not state a context window on the card.",
      "url": "https://aigearwatch.com/can-it-run/#mistral-small-3-2-24b",
      "requiresGb": {
        "q4": 14.4,
        "q5": 18,
        "q8": 28.8,
        "fp16": 57.6
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506",
          "title": "Mistral Small 3.2 24B Instruct model card",
          "publisher": "Mistral AI / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "gemma-4-26b-a4b",
      "name": "Gemma 4 26B-A4B",
      "family": "Gemma 4",
      "paramsB": 25.2,
      "activeParamsB": 3.8,
      "note": "Mixture of experts: Google lists 25.2B total parameters and 3.8B active, with a 256K context window.",
      "url": "https://aigearwatch.com/can-it-run/#gemma-4-26b-a4b",
      "requiresGb": {
        "q4": 15.1,
        "q5": 18.9,
        "q8": 30.2,
        "fp16": 60.5
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/google/gemma-4-26B-A4B-it",
          "title": "Gemma 4 26B-A4B IT model card",
          "publisher": "Google / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "gemma-3-27b",
      "name": "Gemma 3 27B",
      "family": "Gemma 3",
      "paramsB": 27,
      "activeParamsB": null,
      "note": "Model card lists 27B params, with a 128K input context window.",
      "url": "https://aigearwatch.com/can-it-run/#gemma-3-27b",
      "requiresGb": {
        "q4": 16.2,
        "q5": 20.3,
        "q8": 32.4,
        "fp16": 64.8
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/google/gemma-3-27b-it",
          "title": "Gemma 3 27B IT model card",
          "publisher": "Google / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "qwen3-5-27b",
      "name": "Qwen3.5-27B",
      "family": "Qwen3.5",
      "paramsB": 27,
      "activeParamsB": null,
      "note": "Model card lists 27B parameters and a native 262,144-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-5-27b",
      "requiresGb": {
        "q4": 16.2,
        "q5": 20.3,
        "q8": 32.4,
        "fp16": 64.8
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.5-27B",
          "title": "Qwen3.5-27B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "qwen3-8-27b",
      "name": "Qwen3.8-27B",
      "family": "Qwen3.8",
      "paramsB": 27,
      "activeParamsB": null,
      "note": "Model card lists 27B parameters and a native 262,144-token context window, extensible to 1,000,000 tokens.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-8-27b",
      "requiresGb": {
        "q4": 16.2,
        "q5": 20.3,
        "q8": 32.4,
        "fp16": 64.8
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.8-27B",
          "title": "Qwen3.8-27B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "gemma-4-31b",
      "name": "Gemma 4 31B",
      "family": "Gemma 4",
      "paramsB": 30.7,
      "activeParamsB": null,
      "note": "Google lists 30.7B total parameters and a 256K context window for the dense model.",
      "url": "https://aigearwatch.com/can-it-run/#gemma-4-31b",
      "requiresGb": {
        "q4": 18.4,
        "q5": 23,
        "q8": 36.8,
        "fp16": 73.7
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/google/gemma-4-31B-it",
          "title": "Gemma 4 31B IT model card",
          "publisher": "Google / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "qwen3-5-35b-a3b",
      "name": "Qwen3.5-35B-A3B",
      "family": "Qwen3.5",
      "paramsB": 35,
      "activeParamsB": 3,
      "note": "Mixture of experts: model card lists 35B total parameters and 3B active, with a native 262,144-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-5-35b-a3b",
      "requiresGb": {
        "q4": 21,
        "q5": 26.3,
        "q8": 42,
        "fp16": 84
      },
      "fitsOnTrackedAtQ4": 19,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B",
          "title": "Qwen3.5-35B-A3B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "llama-3-3-70b",
      "name": "Llama 3.3 70B Instruct",
      "family": "Llama 3.3",
      "paramsB": 70,
      "activeParamsB": null,
      "note": "Card table lists 70B; the computed safetensors count reads 71B. We use the nominal 70B.",
      "url": "https://aigearwatch.com/can-it-run/#llama-3-3-70b",
      "requiresGb": {
        "q4": 42,
        "q5": 52.5,
        "q8": 84,
        "fp16": 168
      },
      "fitsOnTrackedAtQ4": 15,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct",
          "title": "Llama-3.3-70B-Instruct model card",
          "publisher": "Meta / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "gpt-oss-120b",
      "name": "gpt-oss-120b",
      "family": "gpt-oss",
      "paramsB": 117,
      "activeParamsB": 5.1,
      "note": "OpenAI: \"117B parameters with 5.1B active parameters\"; designed to fit a single 80GB GPU using MXFP4.",
      "url": "https://aigearwatch.com/can-it-run/#gpt-oss-120b",
      "requiresGb": {
        "q4": 70.2,
        "q5": 87.8,
        "q8": 140.4,
        "fp16": 280.8
      },
      "fitsOnTrackedAtQ4": 11,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/openai/gpt-oss-120b",
          "title": "gpt-oss-120b model card",
          "publisher": "OpenAI / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "qwen3-5-122b-a10b",
      "name": "Qwen3.5-122B-A10B",
      "family": "Qwen3.5",
      "paramsB": 122,
      "activeParamsB": 10,
      "note": "Mixture of experts: model card lists 122B total parameters and 10B active, with a native 262,144-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-5-122b-a10b",
      "requiresGb": {
        "q4": 73.2,
        "q5": 91.5,
        "q8": 146.4,
        "fp16": 292.8
      },
      "fitsOnTrackedAtQ4": 11,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.5-122B-A10B",
          "title": "Qwen3.5-122B-A10B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "qwen3-235b-a22b",
      "name": "Qwen3-235B-A22B",
      "family": "Qwen3",
      "paramsB": 235,
      "activeParamsB": 22,
      "note": "Mixture of experts: \"235B in total and 22B activated\". All weights must still be resident.",
      "url": "https://aigearwatch.com/can-it-run/#qwen3-235b-a22b",
      "requiresGb": {
        "q4": 141,
        "q5": 176.3,
        "q8": 282,
        "fp16": 564
      },
      "fitsOnTrackedAtQ4": 2,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3-235B-A22B",
          "title": "Qwen3-235B-A22B model card",
          "publisher": "Qwen / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "deepseek-v4-flash-0731",
      "name": "DeepSeek-V4-Flash-0731",
      "family": "DeepSeek",
      "paramsB": 284,
      "activeParamsB": 13,
      "note": "Mixture of experts: the DeepSeek-V4 report lists 284B total parameters and 13B active, with a one-million-token context window.",
      "url": "https://aigearwatch.com/can-it-run/#deepseek-v4-flash-0731",
      "requiresGb": {
        "q4": 170.4,
        "q5": 213,
        "q8": 340.8,
        "fp16": 681.6
      },
      "fitsOnTrackedAtQ4": 2,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731",
          "title": "DeepSeek-V4-Flash-0731 model card",
          "publisher": "DeepSeek / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        },
        {
          "url": "https://arxiv.org/abs/2606.19348",
          "title": "DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence",
          "publisher": "DeepSeek AI",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "deepseek-v3",
      "name": "DeepSeek-V3",
      "family": "DeepSeek",
      "paramsB": 671,
      "activeParamsB": 37,
      "note": "\"671B total parameters\" with \"37B activated for each token\". The HF upload totals 685B including the MTP module.",
      "url": "https://aigearwatch.com/can-it-run/#deepseek-v3",
      "requiresGb": {
        "q4": 402.6,
        "q5": 503.3,
        "q8": 805.2,
        "fp16": 1610.4
      },
      "fitsOnTrackedAtQ4": 2,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3",
          "title": "DeepSeek-V3 model card",
          "publisher": "DeepSeek / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    },
    {
      "id": "glm-5-1",
      "name": "GLM-5.1",
      "family": "GLM",
      "paramsB": 744,
      "activeParamsB": 40,
      "note": "Mixture of experts: the official GLM-5 repository lists GLM-5.1 as 744B total parameters with 40B active.",
      "url": "https://aigearwatch.com/can-it-run/#glm-5-1",
      "requiresGb": {
        "q4": 446.4,
        "q5": 558,
        "q8": 892.8,
        "fp16": 1785.6
      },
      "fitsOnTrackedAtQ4": 2,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/zai-org/GLM-5.1",
          "title": "GLM-5.1 model card",
          "publisher": "Z.ai / Hugging Face",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        },
        {
          "url": "https://github.com/zai-org/GLM-5",
          "title": "GLM-5 official repository",
          "publisher": "Z.ai",
          "fetchedAt": "2026-09-15T19:40:00.000-07:00"
        }
      ]
    },
    {
      "id": "kimi-k3",
      "name": "Kimi K3",
      "family": "Kimi",
      "paramsB": 2800,
      "activeParamsB": 104,
      "note": "Mixture of experts: card table reads Total Parameters 2.8T, Activated Parameters 104B, 896 experts with 16 selected per token, 1,048,576 context. All weights must be resident, so nothing we track fits it at any quantization.",
      "url": "https://aigearwatch.com/can-it-run/#kimi-k3",
      "requiresGb": {
        "q4": 1680,
        "q5": 2100,
        "q8": 3360,
        "fp16": 6720
      },
      "fitsOnTrackedAtQ4": 0,
      "trackedProductCount": 19,
      "sources": [
        {
          "url": "https://huggingface.co/moonshotai/Kimi-K3",
          "title": "Kimi-K3 model card",
          "publisher": "Moonshot AI / Hugging Face",
          "fetchedAt": "2026-08-16T00:00:00.000Z"
        }
      ]
    }
  ],
  "modelCount": 25,
  "sourceCount": 63
}