Skip to content

Qwen3.5 122B A10B on M3 Ultra 256 GB: VRAM, speed and fit

Qwen3.5 122B A10B at 8-bit runs on the M3 Ultra 256 GB with 256K of context at about 37 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on M3 Ultra 256 GB · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
195 GB
of 256 GB
Context
256K
auto
Speed
37 tok/s
26 when full
First reply
5.6 s
2K prompt

Where the memory goes

Model
124 GB
Context
6.1 GB
Overhead
837 MB
OS
64.0 GB
Free
61.3 GB
Memory
195 GB of 256 GB
Context
256K tokens · auto
Speed
37 tok/s · 26 when full
First reply
5.6 s for a 2K prompt
Running cost
$3.55/mo at home · $287/mo on a cloud 4× A40

Validation: Validated with conditions

What Qwen3.5 122B A10B on M3 Ultra 256 GB was checked against
CheckTargetResultStatus
Fits in GPU memory192 GB usable131 GB neededPass
Headroom5% free61.3 GB free (32%)Pass
Context for chat16K or more256KPass
Writing speed15 tok/s or more37 tok/sPass
First replyunder 3 s, 2K prompt5.6 sCheck

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
Qwen3.5 122B A10B
Parameters
125B, 10B active
Precision
8-bit (Q8_0)
Context
256K, 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

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

Ollama
ollama run qwen3.5:122b-a10b
llama.cpp
llama-server -hf unsloth/Qwen3.5-122B-A10B-GGUF
vLLM
vllm serve Qwen/Qwen3.5-122B-A10B
MLX (Mac)
mlx_lm.chat --model mlx-community/Qwen3.5-122B-A10B-4bit

Qwen3.5 122B A10B on similar machines

All machines: Qwen3.5 122B A10B 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 Qwen3.5 122B A10B run on M3 Ultra 256 GB?

Qwen3.5 122B A10B at 8-bit runs on the M3 Ultra 256 GB with 256K of context at about 37 tokens a second.

Report VR-15K4YG7, calculated Nov 22, 2026 from published specs, not benchmarked.