Skip to content

MiniMax M2 on M3 Ultra 256 GB: VRAM, speed and fit

MiniMax M2 at 6-bit runs on the M3 Ultra 256 GB with 28K of context at about 38 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on M3 Ultra 256 GB · 6-bit (Q6_K) · Ollama / LM Studio · Chat

GPU memory
246 GB
of 256 GB
Context
28K
auto
Speed
38 tok/s
25 when full
First reply
6.9 s
2K prompt

Where the memory goes

Model
175 GB
Context
6.8 GB
Overhead
743 MB
OS
64.0 GB
Free
9.8 GB
Memory
246 GB of 256 GB
Context
28K tokens · auto
Speed
38 tok/s · 25 when full
First reply
6.9 s for a 2K prompt
Running cost
$3.55/mo at home · $287/mo on a cloud 4× A40

Validation: Not validated

What MiniMax M2 on M3 Ultra 256 GB was checked against
CheckTargetResultStatus
Fits in GPU memory192 GB usable182 GB neededPass
Headroom5% free9.8 GB free (5%)Pass
Context for chat16K or more28KPass
Writing speed15 tok/s or more38 tok/sPass
First replyunder 3 s, 2K prompt6.9 sFails

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
MiniMax M2
Parameters
229B, 11.6B active
Precision
6-bit (Q6_K)
Context
28K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
M3 Ultra 256 GB
GPU can use
192 GB of 256 GB

Worth trying

  • 8-bit cache: Room for 52K context

Running cost

At home
$3.55/mo
162 W at the wall
Renting instead
$287/mo
4× A40 at $2.36/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 unsloth/MiniMax-M2-GGUF
vLLM
vllm serve MiniMaxAI/MiniMax-M2
MLX (Mac)
mlx_lm.chat --model mlx-community/MiniMax-M2-4bit

MiniMax M2 on similar machines

All machines: MiniMax M2 VRAM requirements. Everything that runs here: what can the M3 Ultra 256 GB run?

Other models that run on M3 Ultra 256 GB

Questions

Can MiniMax M2 run on M3 Ultra 256 GB?

MiniMax M2 at 6-bit runs on the M3 Ultra 256 GB with 28K of context at about 38 tokens a second.

Report VR-0DNTO46, calculated Nov 22, 2026 from published specs, not benchmarked.