BAAI models
2 BAAI models from BAAI on Hugging Face, from 2.5B to 9.2B parameters, published by BAAI in F32, with a 8K-token context. The smallest official model, bge-reranker-v2-gemma, needs about 11.2 GB of VRAM at its published precision; the cheapest live fit is V100 at $0.110/hr.
Pick a size
One row per official BAAI size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| bge-reranker-v2-gemma | 2.5B | 11.2 GB | 2.8 GB | 1.4 GB | V100 | $0.110/hr | 0.56 GB |
| bge-reranker-v2.5-gemma2-lightweight | 9.2B | 41.3 GB | 10.3 GB | 5.2 GB | RTX A6000 | $0.363/hr | not published for this architecture |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| bge-reranker-v2-gemma | 2.5B | F32 | 11.2 GB | V100 | $0.110/hr |
| bge-reranker-v2.5-gemma2-lightweight | 9.2B | F32 | 41.3 GB | RTX A6000 | $0.363/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.