{
  "id": "bge-small-en-v1-5",
  "identity": {
    "family": "bge",
    "display_name": "bge-small-en-v1.5",
    "hub_id": "BAAI/bge-small-en-v1.5",
    "api_name": null,
    "revision": "5c38ec7c405ec4b44b94cc5a9bb96e735b38267a",
    "variants": [],
    "license": "MIT",
    "commercial_use": true,
    "commercial_use_note": "MIT weights; no use restrictions.",
    "release_date": "2023-09",
    "paper_url": "https://arxiv.org/abs/2309.07597",
    "repo_url": "https://huggingface.co/BAAI/bge-small-en-v1.5"
  },
  "task": {
    "category": "embedding",
    "modalities_in": [
      "text"
    ],
    "modalities_out": [
      "vector"
    ],
    "best_at": "Small English text vectors (33M params, 384 dims) that a template imports from the Hub by repo and revision, so a namespace runs a model the user picked.",
    "worst_at": "Non-English text and passages over 512 tokens, which are truncated."
  },
  "io": {
    "input": {
      "text": {
        "max_tokens": 512,
        "truncation": true,
        "language": "English"
      },
      "prompts": "none"
    },
    "output": {}
  },
  "mixpeek_wiring": {
    "extractors": [
      {
        "extractor": "bge_small_hf@v1",
        "selected_by": "fixed (a manifest-declared extractor that extends text_extractor@v1 and names this model)"
      }
    ],
    "feature_uris": [
      "mixpeek://bge_small_hf@v1/embedding"
    ],
    "vector_indexes": [],
    "retriever_stages": [
      "feature_search"
    ],
    "composes_with": [
      "a manifest extractor with source.huggingface (repo BAAI/bge-small-en-v1.5, a pinned revision) -> text vectors -> feature_search with a typed query"
    ]
  },
  "run": {
    "status": "plugin",
    "status_reason": "No Serve app or built-in extractor loads it. A manifest that declares it under extractors with source.huggingface imports it into the organization's model registry at apply (the bring-your-own-model template does).",
    "serve_apps": [],
    "runtime": "transformers AutoModel + AutoTokenizer (BertModel), CLS pooling",
    "weights_format": "safetensors",
    "trust_remote_code": false,
    "size_on_disk_mb": 133,
    "params_millions": 33.4,
    "preprocessing": "Tokenize with truncation at 512 tokens.",
    "postprocessing": "CLS token embedding, L2-normalized (384 dims).",
    "serving_shape": "both",
    "batch_size": null,
    "serve_batch": null,
    "scale_to_zero": {
      "min_replicas": null,
      "max_replicas": null,
      "idle_timeout_s": null,
      "memory_gb": null,
      "num_cpus": null,
      "num_gpus": null,
      "cold_start_s": {
        "warm_node_s": null,
        "new_node_s": null
      },
      "cpu_fallback": true
    }
  },
  "cost": {
    "unmeasured": true,
    "measured": null,
    "note": "No committed throughput or cost measurement."
  },
  "evidence": {
    "unmeasured": true,
    "items": []
  },
  "refs": [
    "hf://BAAI/bge-small-en-v1.5@5c38ec7c405ec4b44b94cc5a9bb96e735b38267a"
  ],
  "formats": [
    {
      "ref": "hf://BAAI/bge-small-en-v1.5@5c38ec7c405ec4b44b94cc5a9bb96e735b38267a",
      "format": "safetensors",
      "file": "model.safetensors",
      "loader": "transformers from_pretrained; Mixpeek Hub import",
      "supported_today": true
    },
    {
      "ref": "hf://BAAI/bge-small-en-v1.5@5c38ec7c405ec4b44b94cc5a9bb96e735b38267a",
      "format": "onnx",
      "file": "onnx/model.onnx",
      "loader": "onnxruntime; Mixpeek model upload",
      "supported_today": true
    },
    {
      "ref": "hf://BAAI/bge-small-en-v1.5@5c38ec7c405ec4b44b94cc5a9bb96e735b38267a",
      "format": "pytorch_bin",
      "file": "pytorch_model.bin",
      "loader": "torch.load (pickle); Mixpeek refuses pickle weights",
      "supported_today": false
    }
  ],
  "use_cases": [
    "text-search"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "bge-small-en-v1.5: embedding model",
    "description": "bge-small-en-v1.5 is a text embedding model (MIT). Runs on Mixpeek when a template imports it from Hugging Face.",
    "keywords": [
      "bge-small-en-v1.5",
      "bge",
      "embedding",
      "BAAI/bge-small-en-v1.5",
      "text-search"
    ],
    "faqs": [
      {
        "q": "What is bge-small-en-v1.5 best at?",
        "a": "Small English text vectors (33M params, 384 dims) for search over short documents."
      },
      {
        "q": "What are the limits of bge-small-en-v1.5?",
        "a": "Non-English text, and passages over 512 tokens, which are truncated."
      },
      {
        "q": "Can bge-small-en-v1.5 be used commercially?",
        "a": "Yes. The weights are MIT licensed."
      },
      {
        "q": "How does bge-small-en-v1.5 run on Mixpeek?",
        "a": "A manifest names it under extractors with source.huggingface and a pinned revision, and apply imports it into the organization's model registry. The bring-your-own-model template does this."
      },
      {
        "q": "What does bge-small-en-v1.5 output?",
        "a": "A 384-dimensional vector per text, compared by cosine similarity."
      }
    ]
  },
  "last_verified": "2026-11-20",
  "sources": [
    "https://huggingface.co/BAAI/bge-small-en-v1.5",
    "https://huggingface.co/api/models/BAAI/bge-small-en-v1.5",
    "https://arxiv.org/abs/2309.07597"
  ],
  "templates": [
    "bring-your-own-model"
  ],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 1,
    "counted_at": "2026-11-06T13:42:25Z"
  },
  "facets": {
    "status": "plugin",
    "category": "embedding",
    "family": "bge",
    "license": "MIT",
    "commercial_use": true,
    "modalities_in": [
      "text"
    ],
    "modalities_out": [
      "vector"
    ],
    "use_cases": [
      "text-search"
    ],
    "recommended": false,
    "release_date": "2023-09",
    "params_millions": 33.4,
    "size_on_disk_mb": 133,
    "vector_dims": null,
    "vector_distance": null,
    "weights_format": "safetensors",
    "formats": [
      "onnx",
      "pytorch_bin",
      "safetensors"
    ],
    "loadable_today": true,
    "trust_remote_code": false,
    "cpu_fallback": true,
    "num_gpus": null,
    "memory_gb": null,
    "min_replicas": null,
    "serving_shape": "both",
    "served": false,
    "ref_schemes": [
      "hf"
    ],
    "has_evidence": false,
    "cost_measured": false,
    "template_count": 1,
    "usage_namespaces": null,
    "usage_collections": null,
    "usage_uncounted": true,
    "has_demo": false,
    "rank_score": 2.5
  },
  "rank_score": 2.5
}
