Model weights used by Photo Search — find photos by what's in them (https://photo-search.skillsafe.ai/)

This app runs 2 models on your device. About 111 MB of weights are downloaded from https://models.skillsafe.ai/ on first use and cached by your browser; inference happens locally and nothing you enter is uploaded to load them.

- ResNet-50 ImageNet-1k classification (resnet-50-q8@a16f2914, q8) — 24.6 MB — Apache-2.0
  ResNet-50: Microsoft (He et al.), Apache License 2.0 (https://huggingface.co/microsoft/resnet-50); ONNX export by Xenova.
  Notice: https://huggingface.co/microsoft/resnet-50/blob/main/README.md
  Source: https://huggingface.co/skillsafe-ai/resnet-50/tree/37207401008b29c2b3f600e6ab1a4b5398dc6f6f
- all-MiniLM-L6-v2 sentence embeddings (384-d) (all-minilm-l6-v2@1110a243, fp32) — 86.2 MB — Apache-2.0
  all-MiniLM-L6-v2: sentence-transformers (Nils Reimers et al.), Apache License 2.0. https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2 — the repo's own ONNX export.
  Notice: https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2/blob/1110a243fdf4706b3f48f1d95db1a4f5529b4d41/README.md
  Source: https://huggingface.co/skillsafe-ai/all-minilm-l6-v2/tree/de4aaa96e1660bbc9ee25a083f3feb905af95b1d

Catalogue: https://api.skillsafe.ai/v1/browser-models