Skip to content

MiMo V2.6 Pro on 4× H200 NVL: VRAM, speed and fit

MiMo V2.6 Pro at FP4 + FP8 runs on the 4× H200 NVL with 112K of context at about 71 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on 4× H200 NVL · FP4 + FP8 · Ollama / LM Studio · Chat

GPU memory
536 GB
of 564 GB
Context
112K
auto
Speed
71 tok/s
Very fast
First reply
0.7 s
2K prompt

Where the memory goes

Model
527 GB
Context
5.5 GB
Overhead
3.0 GB
OS
0.0 GB
Free
28.3 GB
Memory
536 GB of 564 GB
Context
112K tokens · auto
Speed
71 tok/s · very fast
First reply
0.7 s for a 2K prompt
Running cost
$33/mo at home · $1,162/mo on a cloud 4× MI300X

Validation: Validated

What MiMo V2.6 Pro on 4× H200 NVL was checked against
CheckTargetResultStatus
Fits in GPU memory564 GB usable536 GB neededPass
Headroom5% free28.3 GB free (5%)Pass
Context for chat16K or more112KPass
Writing speed15 tok/s or more71 tok/sPass
First replyunder 3 s, 2K prompt0.7 sPass

Targets: a reply you can read along with, a first reply the use case tolerates, and 5% of memory left free. Calculated from published specs, not benchmarked.

The setup

Model
MiMo V2.6 Pro
Parameters
1T, 42B active
Precision
FP4 + FP8
Context
112K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
4× H200 NVL
GPU can use
564 GB of 564 GB

Running cost

At home
$33/mo
1520 W at the wall
Renting instead
$1,162/mo
4× MI300X at $9.56/h

4 h a day, power at $0.18 per kWh, cloud at RunPod on-demand prices as of Nov 22, 2026.

How to run it

llama.cpp
llama-server -hf ggml-org/MiMo-V2.6-Pro-RL-GGUF
vLLM
vllm serve XiaomiMiMo/MiMo-V2.6-Pro-RL
MLX (Mac)
mlx_lm.chat --model mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8

MiMo V2.6 Pro on similar machines

All machines: MiMo V2.6 Pro VRAM requirements. Everything that runs here: what can the 4× H200 NVL run?

Other models that run on 4× H200 NVL

Questions

Can MiMo V2.6 Pro run on 4× H200 NVL?

MiMo V2.6 Pro at FP4 + FP8 runs on the 4× H200 NVL with 112K of context at about 71 tokens a second.

Report VR-077HEIN, calculated Nov 22, 2026 from published specs, not benchmarked.