{
  "id": "efficientsam-vits",
  "identity": {
    "family": "efficientsam",
    "display_name": "EfficientSAM ViT-S",
    "hub_id": "yformer/EfficientSAM (efficient_sam_vits.pt)",
    "api_name": null,
    "revision": null,
    "variants": [
      "efficient_sam_vitt.pt (ViT-Tiny)"
    ],
    "license": "Apache-2.0",
    "commercial_use": true,
    "commercial_use_note": "Apache-2.0 weights and code; no use restrictions.",
    "release_date": "2023-12",
    "paper_url": "https://arxiv.org/abs/2312.00863",
    "repo_url": "https://github.com/yformer/EfficientSAM"
  },
  "task": {
    "category": "segmentation",
    "modalities_in": [
      "image"
    ],
    "modalities_out": [
      "masks",
      "scores"
    ],
    "best_at": "SAM-style point and box prompted masks at a fraction of SAM ViT-H cost (masked-image pretraining distillation).",
    "worst_at": "Images only (no video memory); below SAM 2 tiny in quality and tooling."
  },
  "io": {
    "input": {
      "image": {
        "resolution": "1024x1024"
      },
      "prompts": "points and boxes"
    },
    "output": {
      "masks": "binary masks with IoU predictions"
    }
  },
  "mixpeek_wiring": {
    "extractors": [],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [],
    "composes_with": [
      "detector boxes -> EfficientSAM masks -> masked crops -> SigLIP"
    ]
  },
  "run": {
    "status": "not_runnable",
    "status_reason": "Weights ship as .pt pickles (ViT-S zipped) and the model code lives in the GitHub repo; needs a safetensors conversion and a vendored model definition (or the repo's ONNX export) before it can be a plugin. SAM 2.1 tiny covers the same need with transformers support.",
    "serve_apps": [],
    "runtime": "repo build_efficient_sam_vits() / ONNX export",
    "weights_format": "pickle",
    "trust_remote_code": false,
    "size_on_disk_mb": null,
    "params_millions": null,
    "preprocessing": "Resize to 1024, normalize.",
    "postprocessing": "Threshold mask logits, pick by predicted IoU.",
    "serving_shape": "both",
    "batch_size": null,
    "serve_batch": null,
    "scale_to_zero": {
      "min_replicas": 0,
      "max_replicas": 1,
      "idle_timeout_s": 600,
      "memory_gb": 2,
      "num_cpus": 2,
      "num_gpus": 0,
      "cold_start_s": {
        "warm_node_s": null,
        "new_node_s": null,
        "source": null
      },
      "cpu_fallback": true
    }
  },
  "cost": {
    "unmeasured": true,
    "measured": null,
    "note": "No committed measurement; plan row 3 measures it in the standalone image."
  },
  "evidence": {
    "unmeasured": true,
    "items": []
  },
  "refs": [
    "github://yformer/EfficientSAM/weights/efficient_sam_vits.pt.zip"
  ],
  "formats": [
    {
      "ref": "github://yformer/EfficientSAM/weights/efficient_sam_vits.pt.zip",
      "format": "pytorch_pt",
      "file": "efficient_sam_vits.pt.zip",
      "loader": "torch.load (pickle); Mixpeek refuses pickle weights (shared/security/torch_pickle.py)",
      "supported_today": false
    }
  ],
  "use_cases": [
    "segmentation",
    "object-detection"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "EfficientSAM ViT-S: segmentation model",
    "description": "EfficientSAM ViT-S is a image segmentation model (Apache-2.0). Not runnable on Mixpeek today.",
    "keywords": [
      "EfficientSAM ViT-S",
      "efficientsam",
      "segmentation",
      "yformer/EfficientSAM (efficient_sam_vits.pt)",
      "object-detection"
    ],
    "faqs": [
      {
        "q": "What is EfficientSAM ViT-S best at?",
        "a": "SAM-style point and box prompted masks at a fraction of SAM ViT-H cost (masked-image pretraining distillation)."
      },
      {
        "q": "What are the limits of EfficientSAM ViT-S?",
        "a": "Images only (no video memory); below SAM 2 tiny in quality and tooling."
      },
      {
        "q": "Can EfficientSAM ViT-S be used commercially?",
        "a": "Yes. Apache-2.0 weights and code; no use restrictions."
      },
      {
        "q": "How does EfficientSAM ViT-S run on Mixpeek?",
        "a": "Weights ship as .pt pickles (ViT-S zipped) and the model code lives in the GitHub repo; needs a safetensors conversion and a vendored model definition (or the repo's ONNX export) before it can be a plugin. SAM 2.1 tiny covers the same need with transformers support."
      }
    ]
  },
  "last_verified": "2026-11-05",
  "sources": [
    "https://github.com/yformer/EfficientSAM",
    "https://arxiv.org/abs/2312.00863"
  ],
  "registry": null,
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": null,
    "source": null
  },
  "facets": {
    "status": "not_runnable",
    "category": "segmentation",
    "family": "efficientsam",
    "license": "Apache-2.0",
    "commercial_use": true,
    "modalities_in": [
      "image"
    ],
    "modalities_out": [
      "masks",
      "scores"
    ],
    "use_cases": [
      "segmentation",
      "object-detection"
    ],
    "recommended": false,
    "release_date": "2023-12",
    "params_millions": null,
    "size_on_disk_mb": null,
    "vector_dims": null,
    "vector_distance": null,
    "weights_format": "pickle",
    "formats": [
      "pytorch_pt"
    ],
    "loadable_today": false,
    "trust_remote_code": false,
    "cpu_fallback": true,
    "num_gpus": 0,
    "memory_gb": 2,
    "min_replicas": 0,
    "serving_shape": "both",
    "served": false,
    "ref_schemes": [
      "github"
    ],
    "has_evidence": false,
    "cost_measured": false,
    "template_count": 0,
    "usage_namespaces": null,
    "usage_collections": null,
    "usage_uncounted": true,
    "has_demo": false,
    "rank_score": 0
  },
  "rank_score": 0
}