Skip to content

MiniMax M3 on DGX Station GB300: VRAM, speed and fit

MiniMax M3 at 3-bit runs on the DGX Station GB300 with 352K of context at about 171 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on DGX Station GB300 · 3-bit (Q3_K_M) · Ollama / LM Studio · Chat

GPU memory
237 GB
of 252 GB
Context
352K
auto
Speed
171 tok/s
68 when full
First reply
0.4 s
2K prompt

Where the memory goes

Model
194 GB
Context
41.3 GB
Overhead
1.0 GB
OS
0.0 GB
Free
15.3 GB
In RAM
0.0 GB
Memory
237 GB of 252 GB
Context
352K tokens · auto
Speed
171 tok/s · 68 when full
First reply
0.4 s for a 2K prompt
Running cost
$23/mo at home · $574/mo on a cloud 8× A40

Validation: Validated

What MiniMax M3 on DGX Station GB300 was checked against
CheckTargetResultStatus
Fits in GPU memory252 GB usable237 GB neededPass
Headroom5% free15.3 GB free (6%)Pass
Context for chat16K or more352KPass
Writing speed15 tok/s or more171 tok/sPass
First replyunder 3 s, 2K prompt0.4 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
MiniMax M3
Parameters
427B, 23B active
Precision
3-bit (Q3_K_M)
Context
352K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
DGX Station GB300
GPU can use
252 GB of 252 GB
Spill space
480 GB LPDDR5X on the Grace CPU

Running cost

At home
$23/mo
1040 W at the wall
Renting instead
$574/mo
8× A40 at $4.72/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-M3-GGUF
vLLM
vllm serve MiniMaxAI/MiniMax-M3
MLX (Mac)
mlx_lm.chat --model mlx-community/MiniMax-M3-4bit

MiniMax M3 on similar machines

All machines: MiniMax M3 VRAM requirements. Everything that runs here: what can the DGX Station GB300 run?

Other models that run on DGX Station GB300

Questions

Can MiniMax M3 run on DGX Station GB300?

MiniMax M3 at 3-bit runs on the DGX Station GB300 with 352K of context at about 171 tokens a second.

Report VR-1U0EZF4, calculated Nov 22, 2026 from published specs, not benchmarked.