MiMo V2 models
4 MiMo V2 models from Xiaomi MiMo on Hugging Face, from 9.4B to 1023.2B parameters, published by Xiaomi MiMo in F8_E4M3, with contexts from 256K to 1M tokens. The smallest official model, MiMo-V2-Flash, needs about 346 GB of VRAM at its published precision; the cheapest live fit is AMD MI300X at $5.26/hr.
Pick a size
One row per official MiMo V2 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| MiMo-V2-Flash | 309.8B | 346 GB | – | 173 GB | AMD MI300X × 2 | $5.26/hr | not published for this architecture |
| MiMo-V2.5 | 310.8B | 347 GB | – | 174 GB | AMD MI300X × 2 | $5.26/hr | not published for this architecture |
| MiMo-V2.5-Pro | 1023.2B | 1144 GB | – | 572 GB | AMD MI300X × 6 | $15.77/hr | not published for this architecture |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (3)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| MiMo-V2-Flash | 309.8B | F8_E4M3 | 346 GB | AMD MI300X × 2 | $5.26/hr |
| MiMo-V2.5 | 310.8B | F8_E4M3 | 347 GB | AMD MI300X × 2 | $5.26/hr |
| MiMo-V2.5-Pro | 1023.2B | F8_E4M3 | 1144 GB | AMD MI300X × 6 | $15.77/hr |
Fine-tunes and community models (1)
| 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 |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. 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.
More from Xiaomi MiMo
- MiMo (2 models)