MiMo V2 models
4 MiMo V2 models on Hugging Face, from 9.4B to 1023.2B parameters. At the precision each one is published in, the smallest needs about 21.0 GB of VRAM (MiMo-V2.6-Distill-Qwen-9B, cheapest live fit: RTX A5000) and the largest about 1144 GB (MiMo-V2.5-Pro). The cheapest way to run MiMo-V2.6-Distill-Qwen-9B is $0.176/hr.
MiMo V2 models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| MiMo-V2.6-Distill-Qwen-9B | 9.4B | BF16 | 21.0 GB | RTX A5000 | $0.176/hr |
| MiMo-V2-Flash | 309.8B | F8_E4M3 | 346 GB | RTX 4090 × 8 | $3.53/hr |
| MiMo-V2.5 | 310.8B | F8_E4M3 | 347 GB | RTX 4090 × 8 | $3.53/hr |
| MiMo-V2.5-Pro | 1023.2B | F8_E4M3 | 1144 GB | No live fit | – |
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.
