google models
2 google models on Hugging Face, from 105M to 25.8B parameters. At the precision each one is published in, the smallest needs about 0.5 GB of VRAM (medasr, cheapest live fit: V100) and the largest about 57.7 GB (diffusiongemma-26B-A4B-it, cheapest live fit: A100). The cheapest way to run medasr is $0.088/hr.
google models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| medasr | 105M | F32 | 0.5 GB | V100 | $0.088/hr |
| diffusiongemma-26B-A4B-it | 25.8B | BF16 | 57.7 GB | A100 | $1.31/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.
