Yes — with 156.6 GB to spare

Qwen3.5 397B-A17B at Q4_K_M fits your M3 Ultra · 512 GB entirely in unified memory at 8K context, at an estimated 22 tokens per second. There is room for its full 256K window.

Fully in unified memory 8K context Q4_K_M · 226.6 GB Apache 2.0 Released 16 Feb 2026 Vision Not in the Ollama library

Flagship-class at 17B active. A 256 GB Mac Studio or a multi-GPU box at Q4.

What hardware do I need for Qwen3.5 397B-A17B? →

The VRAM budget

weights 226.6 GB

Weights 226.6 GB KV cache @ 8K 0.23 GB Runtime overhead 0.6 GB Free 156.6 GB of 384.0 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
Q8_0	398.8 GB	399.6 GB	—	~12	−0.1% ppl	15.6 GB over
Q6_K	307.8 GB	308.6 GB	256K	16	−0.4% ppl	Long context
Q5_K_M	266.0 GB	266.8 GB	256K	19	−0.8% ppl	Long context
Q4_K_M	226.6 GB	227.4 GB	256K	22	−1.9% ppl	Recommended
Q3_K_M	183.4 GB	184.3 GB	256K	27	−5.4% ppl	Long context
Q2_K	157.2 GB	158.0 GB	256K	31	−15% ppl	Long context

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. Only 15 of its 60 blocks keep a per-token KV cache; the rest are linear-attention, Mamba or convolution blocks with a fixed-size state.

How to run it

llama.cpp LM Studio MLX

terminal

$ pip install mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen3.5-397B-A17B-4bit \
    --max-tokens 512 --prompt "Hello"

Apple's own array framework. The fastest path on Apple Silicon. More on MLX.

01Download is 226.6 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.

02macOS caps what the GPU may wire down at about 384.0 GB of 512 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.

03There is room to go to the model's full 256K context on this card.

See all models for this rig Compare with another model