Downloads · 30 days
6
19% of all-time downloads
Omni-Embed-Mini-0.9B-onnx
k-m-irfan/Omni-Embed-Mini-0.9B-onnx
Omni-Embed-Mini-0.9B-onnx is a feature extraction model from k-m-irfan. Use it when you need embeddings to search or compare text. It is set up for onnx. The card lists the license as apache-2.0.
ONNX export of the Omni-Embed Mini 0.9B multimodal embedding model, built to run in a browser on WebGPU. One model embeds text, images, video, speech, general audio and document pages into a single 1024-dimensional sp…
All-time downloads
31
Public
Repo size
1.9 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.data1.9 GB · 99%
At a glance
- Task
- Feature Extraction
- Library
- onnx
- License
- apache-2.0
- Model type
- omni_embed
- Access
- Public
- Created
- Sep 8, 2026
- Updated
- Sep 8, 2026
- SHA
- abdc9740
Model card
From the Hugging Face model README
Omni-Embed Mini 0.9B — ONNX
ONNX export of the Omni-Embed Mini 0.9B multimodal embedding model, built to run in a browser on WebGPU. One model embeds text, images, video, speech, general audio and document pages into a single 1024-dimensional space.
This is a six-graph bundle driven from JavaScript, not a transformers.js
architecture: the splice, the pooling and two of the vision tower's three inputs
are computed by the caller. See manifest.json for the preconditions a caller
cannot read off the graphs.
| Component | Graph | Notes |
|---|---|---|
| backbone | backbone/model.onnx | takes inputs_embeds; pooling is applied outside the graph |
| embedding table | backbone/embed_tokens.onnx | read row-wise off disk rather than held in a session |
| vision | vision_encoder/model.onnx | resolution-independent; one image per call |
| audio | whisper_encoder/, dasheng_encoder/ | mel front end is computed caller-side |
| projectors | projectors/*.onnx | image, video, and one per audio encoder |
Every component was parity-checked against its PyTorch reference before export;
*/conversion_metadata.json and */parity_report.json carry the measured
numbers.
Preprocessing is part of the model
A still image is squared to 224 with PIL bilinear, then the processor takes it to 256 with a Pillow-compatible bicubic. At 64 merged visual tokens this is not a detail: substituting a browser canvas resize moves the final embedding by cos 0.734. Any client that does not reproduce that chain is a different model, and its vectors must not be mixed into an index built with this one.
Demo
Built for a multimodal retrieval demo whose document index is computed offline in PyTorch while queries are embedded in-browser — the two halves share a space only because of the preprocessing note above.
- Task
- Feature Extraction
- Library
- onnx
- Type
- omni_embed
- License
- apache-2.0
- Created
- Sep 8, 2026
- Updated
- Sep 8, 2026