What GPU do you need to run MiMo V2.5 Pro?
1020B params 1M max context Reasoning
MiMo V2.5 Pro doesn't fit on any single GPU in our database. Try multiple GPUs or system RAM offloading in the full tool.
MMLU 89.4 SWE-bench Verified 78.9
VRAM needed per quantization
| Quantization | Weights | 16K context | 32K context | Full 1M |
|---|---|---|---|---|
| Q4_K_M | 630.0 GB | 631.6 GB | 633.2 GB | 732.4 GB |
| Q8_0 | 1090.0 GB | 1091.6 GB | 1093.2 GB | 1192.4 GB |
Weights plus fp16 KV cache. Leave ~1 GB headroom for the inference engine.
GPUs that can run MiMo V2.5 Pro
None on its own — open the tool to try multiple GPUs or system RAM offloading.