LLMs · Xiaomi
MiMo V2.6 Pro (1T-A42B)
- Parameters
- 1,024B (42.0B active)
- Architecture
- Mixture of experts
- Max context
- 1,024K tokens
- License
- mit
- Released
- 2026-09
How to run it
Quantized files · GGUF
Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.
huggingface.co/ggml-org/MiMo-V2.6-Pro-RL-GGUF (opens in a new tab)
ollama run hf.co/ggml-org/MiMo-V2.6-Pro-RL-GGUFOriginal weights · Hugging Face
Full-precision files from Xiaomi, for fine-tuning, vLLM, Transformers.
huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL (opens in a new tab)
Memory needed
Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.
| Quantization | Quality | Weights | Total at 8K context | Total at 32K context | Download |
|---|---|---|---|---|---|
Q2_K |
Low | 410 GB | 411 GB | 413 GB | GGUF · part 1 of 7 (opens in a new tab) |
Q3_K_M |
Low | 502 GB | 503 GB | 505 GB | GGUF files (opens in a new tab) |
Q4_K_M |
Good | 625 GB | 626 GB | 628 GB | GGUF (Q4_0) (opens in a new tab) |
Q5_K_M |
High | 727 GB | 728 GB | 730 GB | GGUF files (opens in a new tab) |
Q6_K |
High | 840 GB | 841 GB | 843 GB | GGUF files (opens in a new tab) |
Q8_0 |
Very high | 1,086 GB | 1,087 GB | 1,088 GB | GGUF (opens in a new tab) |
F16 |
Full | 2,048 GB | 2,050 GB | 2,051 GB | GGUF (opens in a new tab) Original weights (opens in a new tab) |
Best hardware for MiMo V2.6 Pro (1T-A42B)
| Grade · points | Hardware | Quant | Memory | Speed | Fit |
|---|---|---|---|---|---|
| 59 ptsGood (B) · 59 of 100 points: Model quality 31.2/45, Memory fit 4.0/25, Speed 9.0/15, Memory use 15.0/15 | Apple M3 Ultra (512 GB) | Q2_K | 442 GB / 384 GB | 14.4 tok/s | Heavy offload |
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.