openbmb models
2 openbmb models on Hugging Face, from 2.3B to 9.5B parameters. At the precision each one is published in, the smallest needs about 5.1 GB of VRAM (VoxCPM2, cheapest live fit: RTX 3060) and the largest about 21.2 GB (MiniCPM-SALA, cheapest live fit: RTX A5000). The cheapest way to run VoxCPM2 is $0.110/hr.
openbmb models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| VoxCPM2 | 2.3B | BF16 | 5.1 GB | RTX 3060 | $0.110/hr |
| MiniCPM-SALA | 9.5B | BF16 | 21.2 GB | RTX A5000 | $0.176/hr |
VRAM is for the precision the model is published in, with the same overhead and an 8,192-token context assumed on every page; 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.
