Model reference · open weights
bge-reranker-m3-en-ru is an open-weight embedding model from qilowoq, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
Model for English and Russian This is a truncated version of BAAI/bge-reranker-v2-m3. This model has only English and Russian tokens left in the vocabulary. Thus making it 1.5 smaller than the original model while producing the same embeddings. The model has been truncated in this notebook. FAQ Generate Scores for text Citation If you find this repository useful, please consider giving a star and citation
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | qilowoq |
|---|---|
| Type | Embedding models |
| Parameters (lead) | 375M |
| Context | 8,194 tokens |
| Variants | 1 |
| Runs with | sentence-transformers |
| Based on | BAAI/bge-reranker-v2-m3 |
| Released | 2024-09-05 |
| Popularity | 36k downloads / month |
| Likes | 15 |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| bge-reranker-v2-m3-en-ru | 375M | BF16 | ~0.9 GB | ✓ | Weights ↗ |
Using it via the API
Once AxForge deploys bge-reranker-m3-en-ru for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (bge-reranker-m3-en-ru below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/embeddings \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"bge-reranker-m3-en-ru","input":"text to embed"}'
Licence
Open weights under mit — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Explore