Downloads · 30 days
124
31% of all-time downloads
metaclip-2-worldwide-b32-ONNX
bep256/metaclip-2-worldwide-b32-ONNX
metaclip-2-worldwide-b32-ONNX is a machine learning model from bep256. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers.js. The card lists the license as cc-by-nc-4.0.
facebook/metaclip-2-worldwide-b32 with ONNX weights.
All-time downloads
395
Public
Repo size
4.8 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.onnx_data2.4 GB · 97%
At a glance
- Library
- transformers.js
- License
- cc-by-nc-4.0
- Access
- Public
- Created
- Sep 14, 2026
- Updated
- Oct 4, 2026
- SHA
- 4c7dec26
Base models
Model card
From the Hugging Face model README
facebook/metaclip-2-worldwide-b32 with ONNX weights.
External weights
Download each ONNX graph together with its matching .onnx_data file, and load the graph by filename. Learned weights, preprocessing, tokenizer, and embedding space are unchanged from the previous export. Memory savings depend on the execution provider and its weight packing.
image.onnx keeps the ordinary embedding output. image_patched.onnx additionally exposes nicegal_patch_embeds using ClearCLIP query-query attention; both share image.onnx_data. Patch features are experimental localization signals, not segmentation probabilities. Select the standard graph if you only need embeddings.
The _patched models were modified with ClearCLIP.
- Library
- transformers.js
- License
- cc-by-nc-4.0
- Created
- Sep 14, 2026
- Updated
- Oct 4, 2026