MiMo-V2.6-Pro — VRAM, speed & local setup
Written by Jakub Rusinowski · Last updated
A 1.02T-parameter sparse MoE (384 routed experts, 8 active) with 42B parameters active per token, a 681M-parameter vision encoder and an audio encoder, and a 5-layer multi-token-prediction drafter. 1M-token context, MIT licence. The published weights are the RL-tuned checkpoint, MiMo-V2.6-Pro-RL. Sized at Q4_K_M it needs far more memory than any single machine in our catalogue.
MiMo-V2.6-Pro needs about 619 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
VRAM and speed by quantization
Quoted against NVIDIA RTX 4090 (24 GB). Weights plus overhead only: this model's architecture is unpublished, so no KV cache is added.
| Quant | Memory | VRAM | Speed (est.) | Fit |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 337.5 GB | — | Too big | |
| Q3_K_M 3.41 bpw | 437.4 GB | — | Too big | |
| Q4_K_M 4.83 bpw | 619.2 GB | — | Too big | |
| Q5_K_M 5.67 bpw | 726.7 GB | — | Too big | |
| Q6_K 6.56 bpw | 840.6 GB | — | Too big | |
| Q8_0 8.50 bpw | 1089 GB | — | Too big | |
| F16 16.00 bpw | 2049.2 GB | — | Too big |
Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. MiMo-V2.6-Pro VRAM calculator →
Get MiMo-V2.6-Pro running
MiMo-V2.6-Pro is not practical on a single consumer GPU: it needs about 619.2 GB at Q4_K_M, more than any single device in our catalogue holds with headroom. How multi-GPU inference works, or rent a multi-GPU server by the hour:
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
How to run MiMo-V2.6-Pro
No Ollama library entry. Xiaomi's documented serving routes are SGLang and vLLM.
Specifications
Verified — Checked against the primary source — the model card or the vendor spec page — and corroborated by a second independent source.
- Parameters
- 1.02 Trillion (42B active)
- Context window
- 1,048,576
- Architecture
- Sparse MoE (384 experts, 8 active), hybrid sliding-window attention; vision + audio encoders
- Provider
- Xiaomi
- Licence
- MIT
- Availability
- Weights downloadable
- Practical deployment
- Private cloud / datacenter cluster
- Specified at
- Q4_K_M
- System RAM
- 1024 GB
- Record updated
- 2026-10-13
- Runs with
- SGLang: supported vLLM: supported Ollama: not supported
Commercial use permitted. No usage restrictions beyond attribution.
See also MiMo-V2.5-Pro 1T
Quality and use cases
Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.
Independent evaluation — not vendor-claimed and not measured by us
| Benchmark | Score | Provenance |
|---|---|---|
| AA Intelligence Index v4.3.2 | 46 | independent · Artificial Analysis · snapshot 2026-10-13 |
MiMo-V2.6-Pro — frequently asked questions
How much VRAM does MiMo-V2.6-Pro need?
About 619 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does MiMo-V2.6-Pro run on an RTX 4090 (24 GB)?
No. MiMo-V2.6-Pro needs about 619 GB at Q4_K_M, more than a single RTX 4090's 24 GB. It needs a larger card, several GPUs, or Apple Silicon with enough unified memory — or it runs with part of the weights offloaded to system RAM, which is much slower.
How do I run MiMo-V2.6-Pro locally?
No Ollama library entry. Xiaomi's documented serving routes are SGLang and vLLM. Running the published tag would send your prompts to a hosted GPU rather than your own machine.