K-intelligence models
2 K-intelligence models on Hugging Face, from 2.3B to 11.5B parameters. At the precision each one is published in, the smallest needs about 5.2 GB of VRAM (Midm-2.0-Mini-Instruct, cheapest live fit: RTX 3060) and the largest about 25.8 GB (Midm-2.0-Base-Instruct, cheapest live fit: RTX 4080 Super). The cheapest way to run Midm-2.0-Mini-Instruct is $0.110/hr.
K-intelligence models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Midm-2.0-Mini-Instruct | 2.3B | BF16 | 5.2 GB | RTX 3060 | $0.110/hr |
| Midm-2.0-Base-Instruct | 11.5B | BF16 | 25.8 GB | RTX 4080 Super | $0.338/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.
