Bge m3
Model Overview¶
BGE-M3 is distinguished for its versatility in Multi-Functionality, Multi-Linguality, and Multi-Granularity.
Multi-Functionality: It can simultaneously perform the three common retrieval functionalities of embedding model: dense retrieval, multi-vector retrieval, and sparse retrieval. Multi-Linguality: It can support more than 100 working languages. Multi-Granularity: It is able to process inputs of different granularities, spanning from short sentences to long documents of up to 8192 tokens.
- Model Source: BAAI/bge-m3
- License: MIT License
QPC Configurations¶
| Batch Size | SEQUENCE LENGTH | CORES | OLS | Generated URL | Download |
|---|---|---|---|---|---|
| 1 | 512 | 2 | default | https://dc00tk1pxen80.cloudfront.net/SDK1.19.6/BAAI/bge-m3/compiled-bin-fp16-B1-C2-A7-best-throughput.tar.gz | Download |
| 4 | 512 | 2 | default | https://dc00tk1pxen80.cloudfront.net/SDK1.19.6/BAAI/bge-m3/compiled-bin-fp16-B4-C2-A7-best-throughput.tar.gz | Download |