h2oai models
3 h2oai models on Hugging Face, from 514M to 2.2B parameters. At the precision each one is published in, the smallest needs about 1.1 GB of VRAM (h2o-danube3-500m-chat, cheapest live fit: RTX 5060 Ti) and the largest about 4.8 GB (h2ovl-mississippi-2b, cheapest live fit: RTX 5060 Ti). The cheapest way to run h2o-danube3-500m-chat is $0.110/hr.
h2oai models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| h2o-danube3-500m-chat | 514M | BF16 | 1.1 GB | RTX 5060 Ti | $0.110/hr |
| h2ovl-mississippi-800m | 826M | BF16 | 1.8 GB | RTX 5060 Ti | $0.110/hr |
| h2ovl-mississippi-2b | 2.2B | BF16 | 4.8 GB | RTX 5060 Ti | $0.110/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.
