UnfilteredAI models
2 UnfilteredAI models on Hugging Face, from 2.2B to 2.6B parameters. At the precision each one is published in, the smallest needs about 5.0 GB of VRAM (NSFW-flash, cheapest live fit: V100) and the largest about 11.5 GB (NSFW-GEN-ANIME-v2, cheapest live fit: V100). The cheapest way to run NSFW-flash is $0.088/hr.
UnfilteredAI models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| NSFW-flash | 2.2B | F16 | 5.0 GB | V100 | $0.088/hr |
| NSFW-GEN-ANIME-v2 | 2.6B | F32 | 11.5 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.
