Qwen3.5 397B A17B on M3 Ultra 256 GB: VRAM, speed and fit
Qwen3.5 397B A17B at 2-bit runs on the M3 Ultra 256 GB with 256K of context at about 48 tokens a second.
Text generation · Chat · Ollama / LM Studio
Runs comfortably · on M3 Ultra 256 GB · 2-bit (Q2_K) · Ollama / LM Studio · Chat
- GPU memory
- 213 GB
- of 256 GB
- Context
- 256K
- auto
- Speed
- 48 tok/s
- 28 when full
- First reply
- 9.9 s
- 2K prompt
Where the memory goes
- Model
- 141 GB
- Context
- 7.7 GB
- Overhead
- 869 MB
- OS
- 64.0 GB
- Free
- 42.6 GB
- Memory
- 213 GB of 256 GB
- Context
- 256K tokens · auto
- Speed
- 48 tok/s · 28 when full
- First reply
- 9.9 s for a 2K prompt
- Running cost
- $3.55/mo at home · $287/mo on a cloud 4× A40
Validation: Not validated
| Check | Target | Result | Status |
|---|---|---|---|
| Fits in GPU memory | 192 GB usable | 149 GB needed | Pass |
| Headroom | 5% free | 42.6 GB free (22%) | Pass |
| Context for chat | 16K or more | 256K | Pass |
| Writing speed | 15 tok/s or more | 48 tok/s | Pass |
| First reply | under 3 s, 2K prompt | 9.9 s | Fails |
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 397B A17B
- Parameters
- 403B, 17B active
- Precision
- 2-bit (Q2_K)
- 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 21, 2026.
How to run it
- llama.cpp
llama-server -hf unsloth/Qwen3.5-397B-A17B-GGUF- vLLM
vllm serve Qwen/Qwen3.5-397B-A17B- MLX (Mac)
mlx_lm.chat --model mlx-community/Qwen3.5-397B-A17B-4bit- Model card
- huggingface.co/Qwen/Qwen3.5-397B-A17B
Qwen3.5 397B A17B on similar machines
- Mac mini M4 Pro 64 GBToo big
- M4 Max 128 GBRuns
- M5 Max 128 GBRuns
- M3 Ultra 96 GBJust misses
- M3 Ultra 512 GBRuns
- M5 Ultra 512 GBRuns
- Ryzen AI Max+ 395 64 GBToo big
- Ryzen AI Max+ 395 128 GBRuns
All machines: Qwen3.5 397B A17B 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 397B A17B run on M3 Ultra 256 GB?
Qwen3.5 397B A17B at 2-bit runs on the M3 Ultra 256 GB with 256K of context at about 48 tokens a second.
Report VR-0HFL462, calculated Nov 22, 2026 from published specs, not benchmarked.