convaiinnovations models
3 convaiinnovations models on Hugging Face, from 322M to 421M parameters. At the precision each one is published in, the smallest needs about 0.7 GB of VRAM (laya-multilingual, cheapest live fit: V100) and the largest about 0.9 GB (laya, cheapest live fit: V100). The cheapest way to run laya-multilingual is $0.088/hr.
convaiinnovations models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| laya-multilingual | 322M | F16 | 0.7 GB | V100 | $0.088/hr |
| laya | 421M | F16 | 0.9 GB | V100 | $0.088/hr |
| laya-typed-decisions | 421M | F16 | 0.9 GB | V100 | $0.088/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.
