Skip to content

Kimi K2.6 on 8× H200 NVL: VRAM, speed and fit

Kimi K2.6 at INT4 runs on the 8× H200 NVL with 256K of context at about 58 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on 8× H200 NVL · INT4 · Ollama / LM Studio · Chat

GPU memory
577 GB
of 1,128 GB
Context
256K
auto
Speed
58 tok/s
44 when full
First reply
0.3 s
2K prompt

Where the memory goes

Model
554 GB
Context
17.2 GB
Overhead
6.0 GB
OS
0.0 GB
Free
551 GB
Memory
577 GB of 1,128 GB
Context
256K tokens · auto
Speed
58 tok/s · 44 when full
First reply
0.3 s for a 2K prompt
Running cost
$65/mo at home · $1,162/mo on a cloud 4× MI300X

Validation: Validated

What Kimi K2.6 on 8× H200 NVL was checked against
CheckTargetResultStatus
Fits in GPU memory1,128 GB usable577 GB neededPass
Headroom5% free551 GB free (49%)Pass
Context for chat16K or more256KPass
Writing speed15 tok/s or more58 tok/sPass
First replyunder 3 s, 2K prompt0.3 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
Kimi K2.6
Parameters
1T, 32B active
Precision
INT4
Context
256K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
8× H200 NVL
GPU can use
1,128 GB of 1,128 GB

Running cost

At home
$65/mo
2960 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 21, 2026.

How to run it

llama.cpp
llama-server -hf unsloth/Kimi-K2.6-GGUF
vLLM
vllm serve moonshotai/Kimi-K2.6
MLX (Mac)
mlx_lm.chat --model mlx-community/Kimi-K2.6-mlx-DQ3_K_M-q8

Kimi K2.6 on similar machines

All machines: Kimi K2.6 VRAM requirements. Everything that runs here: what can the 8× H200 NVL run?

Other models that run on 8× H200 NVL

Questions

Can Kimi K2.6 run on 8× H200 NVL?

Kimi K2.6 at INT4 runs on the 8× H200 NVL with 256K of context at about 58 tokens a second.

Report VR-0WZFMPQ, calculated Nov 22, 2026 from published specs, not benchmarked.