{
  "count": 18,
  "kind": "model-runner",
  "sorts": [
    "stars",
    "name"
  ],
  "facets": {
    "local": [
      "true"
    ],
    "server": [
      "true"
    ],
    "lowRam": [
      "true"
    ],
    "gpu": [
      "true"
    ]
  },
  "facetCounts": {
    "local": {
      "classified": 13,
      "total": 18,
      "values": {
        "true": 13
      }
    },
    "server": {
      "classified": 4,
      "total": 18,
      "values": {
        "true": 4
      }
    },
    "lowRam": {
      "classified": 1,
      "total": 18,
      "values": {
        "true": 1
      }
    },
    "gpu": {
      "classified": 1,
      "total": 18,
      "values": {
        "true": 1
      }
    }
  },
  "outputs": [
    "json",
    "xml",
    "csv",
    "sse",
    "events"
  ],
  "resources": [
    {
      "slug": "ollama",
      "name": "Ollama",
      "kind": "model-runner",
      "tags": [
        "local-first"
      ],
      "links": [],
      "url": "https://ollama.com",
      "description": "The default local model runner; CLI + OpenAI-compatible API.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "api",
        "cli"
      ],
      "github": {
        "repo": "ollama/ollama",
        "stars": 181243,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "comfyui",
      "name": "ComfyUI",
      "kind": "model-runner",
      "tags": [
        "diffusion"
      ],
      "links": [],
      "url": "https://www.comfy.org",
      "description": "Node-graph image/video generation UI.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "image-gen",
        "video-gen"
      ],
      "github": {
        "repo": "Comfy-Org/ComfyUI",
        "stars": 133828,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "llama-cpp",
      "name": "llama.cpp",
      "kind": "model-runner",
      "tags": [
        "engine"
      ],
      "links": [],
      "url": "https://github.com/ggml-org/llama.cpp",
      "description": "GGUF inference engine underneath most local runners.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "cli",
        "api"
      ],
      "github": {
        "repo": "ggerganov/llama.cpp",
        "stars": 128752,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "vllm",
      "name": "vLLM",
      "kind": "model-runner",
      "tags": [
        "serving",
        "gpu"
      ],
      "links": [],
      "url": "https://github.com/vllm-project/vllm",
      "description": "High-throughput GPU serving for open models.",
      "access": "self-host",
      "pricing": "free",
      "caps": [
        "server",
        "api"
      ],
      "github": {
        "repo": "vllm-project/vllm",
        "stars": 92134,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "server": [
          "true"
        ],
        "gpu": [
          "true"
        ]
      }
    },
    {
      "slug": "whisper-cpp",
      "name": "whisper.cpp",
      "kind": "model-runner",
      "tags": [],
      "links": [],
      "url": "https://github.com/ggerganov/whisper.cpp",
      "description": "Local Whisper inference.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "stt"
      ],
      "github": {
        "repo": "ggerganov/whisper.cpp",
        "stars": 53770,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "localai",
      "name": "LocalAI",
      "kind": "model-runner",
      "tags": [
        "openai-compat"
      ],
      "links": [],
      "url": "https://localai.io",
      "description": "Self-hosted OpenAI-compatible API.",
      "access": "self-host",
      "pricing": "free",
      "caps": [
        "server",
        "api"
      ],
      "github": {
        "repo": "mudler/LocalAI",
        "stars": 49164,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "server": [
          "true"
        ]
      }
    },
    {
      "slug": "text-gen-webui",
      "name": "text-generation-webui",
      "kind": "model-runner",
      "tags": [],
      "links": [],
      "url": "https://github.com/oobabooga/text-generation-webui",
      "description": "Feature-rich local web UI for models.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "web"
      ],
      "github": {
        "repo": "oobabooga/text-generation-webui",
        "stars": 47685,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "exo",
      "name": "exo",
      "kind": "model-runner",
      "tags": [
        "distributed"
      ],
      "links": [],
      "url": "https://github.com/exo-explore/exo",
      "description": "Cluster everyday devices into a GPU pool.",
      "access": "self-host",
      "pricing": "free",
      "caps": [
        "cluster",
        "local"
      ],
      "github": {
        "repo": "exo-explore/exo",
        "stars": 47514,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "jan",
      "name": "Jan",
      "kind": "model-runner",
      "tags": [
        "desktop"
      ],
      "links": [],
      "url": "https://jan.ai",
      "description": "Open-source local AI desktop app.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "desktop"
      ],
      "github": {
        "repo": "menloresearch/jan",
        "stars": 44558,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "airllm",
      "name": "AirLLM",
      "kind": "model-runner",
      "tags": [
        "low-resource"
      ],
      "links": [],
      "url": "https://github.com/lyogavin/airllm",
      "description": "Run 70B models on 4GB RAM via layer-wise inference.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "low-ram"
      ],
      "github": {
        "repo": "lyogavin/airllm",
        "stars": 34534,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ],
        "lowRam": [
          "true"
        ]
      }
    },
    {
      "slug": "tabby",
      "name": "Tabby",
      "kind": "model-runner",
      "tags": [
        "coding"
      ],
      "links": [],
      "url": "https://tabby.tabbyml.com",
      "description": "Self-hosted coding assistant.",
      "access": "self-host",
      "pricing": "free",
      "caps": [
        "server",
        "code"
      ],
      "github": {
        "repo": "TabbyML/tabby",
        "stars": 33881,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "server": [
          "true"
        ]
      }
    },
    {
      "slug": "invokeai",
      "name": "InvokeAI",
      "kind": "model-runner",
      "tags": [],
      "links": [],
      "url": "https://invoke-ai.github.io",
      "description": "Visual creative engine for diffusion.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "image-gen"
      ],
      "github": {
        "repo": "invoke-ai/InvokeAI",
        "stars": 28239,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "mlc-llm",
      "name": "MLC LLM",
      "kind": "model-runner",
      "tags": [
        "edge",
        "webgpu"
      ],
      "links": [],
      "url": "https://llm.mlc.ai",
      "description": "Compile models for every device incl. phones/browser.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "edge"
      ],
      "github": {
        "repo": "mlc-ai/mlc-llm",
        "stars": 23168,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "koboldcpp",
      "name": "KoboldCpp",
      "kind": "model-runner",
      "tags": [],
      "links": [],
      "url": "https://github.com/LostRuins/koboldcpp",
      "description": "Single-binary local runner.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "api"
      ],
      "github": {
        "repo": "LostRuins/koboldcpp",
        "stars": 11793,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "tgi",
      "name": "Text Generation Inference",
      "kind": "model-runner",
      "tags": [],
      "links": [],
      "url": "https://github.com/huggingface/text-generation-inference",
      "description": "HF's production serving stack.",
      "access": "self-host",
      "pricing": "free",
      "caps": [
        "server",
        "api"
      ],
      "github": {
        "repo": "huggingface/text-generation-inference",
        "stars": 10886,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "server": [
          "true"
        ]
      }
    },
    {
      "slug": "hyperspace-pods",
      "name": "Hyperspace Pods",
      "kind": "model-runner",
      "tags": [
        "decentralized"
      ],
      "links": [],
      "url": "https://hyperspaceai.agi",
      "description": "Decentralized inference pods network.",
      "access": "network",
      "pricing": "free",
      "caps": [
        "distributed",
        "network"
      ],
      "github": {
        "repo": "hyperspaceai/agi",
        "stars": 2054,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {}
    },
    {
      "slug": "lm-studio",
      "name": "LM Studio",
      "kind": "model-runner",
      "tags": [
        "desktop"
      ],
      "links": [],
      "url": "https://lmstudio.ai",
      "description": "Polished local model desktop app + local server.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local",
        "desktop",
        "api"
      ],
      "github": {
        "repo": "lmstudio-ai/lmstudio.js",
        "stars": 1781,
        "starsAsOf": "2026-09-19"
      },
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    },
    {
      "slug": "colibri-ai",
      "name": "Colibri.ai",
      "kind": "model-runner",
      "tags": [
        "lightweight"
      ],
      "links": [],
      "url": "https://github.com/nickr-s/Colibri",
      "description": "Lightweight local inference toolkit.",
      "access": "local",
      "pricing": "free",
      "caps": [
        "local"
      ],
      "github": null,
      "kinds": [
        "model-runner"
      ],
      "facets": {
        "local": [
          "true"
        ]
      }
    }
  ]
}