Skip to content

Laguna M.1 on M3 Ultra 256 GB: VRAM, speed and fit

Laguna M.1 at 6-bit runs on the M3 Ultra 256 GB with 32K of context at about 22 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
32K
auto
Speed
22 tok/s
16 when full
First reply
15 s
2K prompt

Where the memory goes

Model
172 GB
Context
8.8 GB
Overhead
580 MB
OS
64.0 GB
Free
10.2 GB
Memory
246 GB of 256 GB
Context
32K tokens · auto
Speed
22 tok/s · 16 when full
First reply
15 s for a 2K prompt
Running cost
$3.55/mo at home · $287/mo on a cloud 4× A40

Validation: Not validated

What Laguna M.1 on M3 Ultra 256 GB was checked against
CheckTargetResultStatus
Fits in GPU memory192 GB usable182 GB neededPass
Headroom5% free10.2 GB free (5%)Pass
Context for chat16K or more32KPass
Writing speed15 tok/s or more22 tok/sPass
First replyunder 3 s, 2K prompt15 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
Laguna M.1
Parameters
226B, 23.5B active
Precision
6-bit (Q6_K)
Context
32K, 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 64K 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 sigargv/Laguna-M.1-GGUF
vLLM
vllm serve poolside/Laguna-M.1
MLX (Mac)
mlx_lm.chat --model mlx-community/Laguna-M.1-4bit

Laguna M.1 on similar machines

All machines: Laguna M.1 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 Laguna M.1 run on M3 Ultra 256 GB?

Laguna M.1 at 6-bit runs on the M3 Ultra 256 GB with 32K of context at about 22 tokens a second.

Report VR-15QLN3Y, calculated Nov 22, 2026 from published specs, not benchmarked.