| --- |
| license: other |
| license_name: mixed-permissive |
| license_link: https://huggingface.co/umbml/arca-models/blob/main/LICENSES.md |
| library_name: onnx |
| tags: |
| - onnx |
| - coreml |
| - clip |
| - face-recognition |
| - mirror |
| --- |
| |
| # arca-models |
|
|
| Pinned ONNX mirrors of third-party models used for **on-device** inference in |
| [Arca](https://arca.umbml.com). Nothing here is trained by us — every file is a |
| byte-for-byte copy of an upstream release, re-hosted so that: |
|
|
| 1. **Downloads are immutable.** Arca pins each file by commit SHA and verifies a |
| SHA-256 before use. Fetching upstream `main` gave us a moving target: two of |
| our source URLs 404'd after upstream repos were restructured, silently |
| breaking on-device features with nothing able to detect the drift. |
| 2. **Upstream repos aren't used as a CDN.** Pulling a 261 MB file from a |
| `github.com/.../raw/main/` URL on every client install is not what that |
| endpoint is for. |
|
|
| If you want these models, please prefer the original sources below — they are |
| the authoritative, maintained copies. |
|
|
| ## Contents |
|
|
| Every file here is live in Arca. Nothing is staged "for later" — an unused model |
| in a mirror is just a file nobody can explain a year from now. |
|
|
| | File | Bytes | Upstream source | License | |
| | --- | --- | --- | --- | |
| | `arcface.onnx` | 261,036,388 | [onnx/models](https://github.com/onnx/models/tree/main/validated/vision/body_analysis/arcface) — `arcfaceresnet100-8.onnx` | Apache-2.0 | |
| | `auraface.mlpackage.zip` | 241,915,367 | [fal/AuraFace-v1](https://huggingface.co/fal/AuraFace-v1) — `glintr100.onnx`, converted to Core ML | Apache-2.0 | |
| | `clip-vit-b32-image.onnx` | 351,686,194 | [Qdrant/clip-ViT-B-32-vision](https://huggingface.co/Qdrant/clip-ViT-B-32-vision) | MIT | |
| | `clip-vit-b32-text.onnx` | 254,102,519 | [Qdrant/clip-ViT-B-32-text](https://huggingface.co/Qdrant/clip-ViT-B-32-text) | MIT | |
| | `bpe_simple_vocab_16e6.txt.gz` | 1,356,917 | [openai/CLIP](https://github.com/openai/CLIP) | MIT | |
|
|
| ### SHA-256 |
|
|
| ``` |
| f3a6bc281e72f88862f5748b53be3d76b3b48f8f1ab1f4a537941bdc4e1b01da arcface.onnx |
| 7c47e081da5a8eb50c7f34faea9cbffa3f4f95a7e08c0c359ad3a4fa7610aaee auraface.mlpackage.zip |
| c68d3d9a200ddd2a8c8a5510b576d4c94d1ae383bf8b36dd8c084f94e1fb4d63 clip-vit-b32-image.onnx |
| 4dbe762b11e36488304471e439cde89da053ad7acaddbf9e096745d142ec8d8b clip-vit-b32-text.onnx |
| 924691ac288e54409236115652ad4aa250f48203de50a9e4722a6ecd48d6804a bpe_simple_vocab_16e6.txt.gz |
| ``` |
|
|
| ## Note on `auraface.mlpackage.zip` |
|
|
| A `.mlpackage` is a directory bundle, not a single file, so it is zipped |
| (`ditto -c -k --sequesterRsrc --keepParent`) for single-URL distribution and |
| single-SHA-256 pinning, matching every other file in this repo. The installing |
| client (`FaceModelInstaller.swift`) unzips it before handing it to |
| `MLModel.compileModel(at:)` — it never compiles the zip directly. |
|
|
| Converted from `glintr100.onnx` (from the same `fal/AuraFace-v1` repo as its |
| sibling `scrfd_10g_bnkps.onnx`/`2d106det.onnx`/`1k3d68.onnx`/`genderage.onnx` — |
| **only `glintr100.onnx` is mirrored here**. The rest of that repo is stock |
| InsightFace detection/auxiliary models with non-commercial upstream terms; |
| `glintr100.onnx` is the one file `fal/AuraFace-v1`'s own `LICENSE.md` (stock |
| Apache-2.0, verified in full, no field-of-use rider) actually covers, and it is |
| the only one Arca needs — detection and landmarks come from the OS's own face |
| API on every platform, never from this mirror. Bridge: ONNX → PyTorch via |
| `onnx2torch` (`coremltools` dropped direct ONNX conversion) → traced → |
| `coremltools.convert(..., convert_to="mlprogram")`. Verified against the |
| original ONNX output on 5 real photos: max abs diff ~5e-6, cosine similarity |
| 1.0 — floating-point noise, not a converter defect. |
|
|
| ## Note on the CLIP pair |
|
|
| `clip-vit-b32-image.onnx` and `clip-vit-b32-text.onnx` come from the **same |
| export family** (both PyTorch 2.3.0 exports of `clip-ViT-B-32`) and therefore |
| share one 512-d embedding space. Mixing an image encoder from one port with a |
| text encoder from another puts the vectors in different spaces and makes cosine |
| similarity meaningless. Treat the two as a single atomic unit. |
|
|
| ## Previously here, removed |
|
|
| Both were removed from `main` by a normal commit, so earlier pinned |
| `resolve/<sha>/` URLs still resolve — nothing that referenced them breaks. |
|
|
| - **`yunet.onnx`** (face detection, MIT) — Arca uses the OS |
| `Windows.Media.FaceAnalysis.FaceDetector` for detection, so this was never |
| loaded. `arcface.onnx` (the 512-d embedder) is the only ONNX face model. |
| - **`yolox_tiny.onnx`** (object detection, Apache-2.0) — mirrored in |
| anticipation of an object-detection feature that doesn't exist yet, then |
| removed rather than left sitting here unused. If it comes back, it will be |
| re-uploaded alongside the code that actually calls it. |
| |
| ## Licensing |
| |
| Each file keeps its upstream license; see the table above and `LICENSES.md`. |
| The `license: other` / `mixed-permissive` marker reflects that this repo |
| aggregates MIT and Apache-2.0 files rather than being under one single license. |
| No copyleft-licensed weights are hosted here — see `LICENSES.md` for why |
| Ultralytics YOLO in particular can never be. |
| |
