IFM models
2 IFM models on Hugging Face, from 1.1B to 37.4B parameters. At the precision each one is published in, the smallest needs about 2.4 GB of VRAM (K2-Type-0.9B, cheapest live fit: RTX 3060) and the largest about 83.7 GB (K2-Horizon-MoVA-36B-A4B, cheapest live fit: RTX PRO 6000). The cheapest way to run K2-Type-0.9B is $0.110/hr.
IFM models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| K2-Type-0.9B | 1.1B | BF16 | 2.4 GB | RTX 3060 | $0.110/hr |
| K2-Horizon-MoVA-36B-A4B | 37.4B | BF16 | 83.7 GB | RTX PRO 6000 | $1.38/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.
