MiMo Models — Hardware Requirements
8 MiMo models from Xiaomi and the community, from the smallest that runs in 6.5 GB of VRAM at Q4 up to 1020B parameters. Every row links to full quantization tables and GPU compatibility.
All MiMo Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| MiMo 7B RL | 7.8B | 7.4 GB | 33K | ||
| MiMo V2.6 Flash RL | 309B | 186.5 GB | 1049K | ||
| MiMo V2.6 Flash MOPD | 309B | 186.5 GB | 1049K | ||
| MiMo v2 Flash | 309.8B | 187.0 GB | 262K | ||
| MiMo V2.5 | 310.8B | 187.6 GB | 1049K | ||
| MiMo V2.6 Pro RL | 1020B | 614.1 GB | 1049K | ||
| MiMo V2.6 Pro MOPD | 1020B | 614.1 GB | 1049K |
Community variants and fine-tunes(2)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| MiMo V2.6 Distill Qwen 9B | 9.4B | 6.5 GB | 262K | ||
| MiMo V2.6 Pro RL UNCENSORED | 1024.2B | 616.6 GB | 1049K |
How MiMo Compares — llmrun Composite Score
MiMo V2.6 Pro RL is the highest-scoring MiMo model with an llmrun Composite Score of 133.3 — #3 among 106 open models. The top proprietary model, Gemini 4 Carbon, scores 158.6. Click a model to see its full benchmark breakdown.
Gemini 4 Carbon · proprietary158.6
GPT 6 Astra · proprietary156.2
Claude Opus 5.5 · proprietary153.0
GPT 6.1 Sol · proprietary151.7
GPT 6.1 Terra · proprietary150.1
Kimi K3.1 Pro142.8
MiMo V2.6 Pro RL133.3
Kimi K3132.7
MiMo V2.6 Flash RL131.9
GLM 5.3130.1
DeepSeek V4.1 Flash129.4
DeepSeek V4 Pro 0813129.2
Step 5 Preview128.0
MiMo V2.5108.3
Frequently Asked Questions
- How much VRAM do I need to run a MiMo model?
- The smallest MiMo model, MiMo V2.6 Distill Qwen 9B, runs from 6.5 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 3.2 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which MiMo models can I run on a 16 GB GPU?
- 2 of 9 MiMo models fit in 16 GB of VRAM at Q4_K_M, including MiMo V2.6 Distill Qwen 9B, MiMo 7B RL.
- What is the most popular MiMo model to run locally?
- MiMo V2.6 Distill Qwen 9B is the most downloaded MiMo model in local-friendly quantized formats. It runs from 6.5 GB of VRAM at Q4_K_M.
- How do MiMo models score on benchmarks?
- MiMo V2.6 Pro RL leads the family with an llmrun Composite Score of 133.3, ranking #3 among 106 open models, while the top proprietary model, Gemini 4 Carbon, scores 158.6. See the comparison chart above for the full standings.