Yes, just — 1.3 GB spare

Qwen3.6 27B at Q4_K_M fits your M3 · 24 GB entirely in unified memory at 8K context, at an estimated 3.6 tokens per second. Past 28K the KV cache pushes it over — quantise the cache to q8_0, or step down a quantisation, to go longer.

Fully in unified memory 8K context Q4_K_M · 15.6 GB Apache 2.0 Released 22 Apr 2026 Vision

The 24 GB coding pick of spring 2026 (77.2 SWE-bench Verified). Same shape as 3.8, one generation behind.

What hardware do I need for Qwen3.6 27B? →

The VRAM budget

weights 15.6 GB

Weights 15.6 GB KV cache @ 8K 0.50 GB Runtime overhead 0.6 GB Free 1.3 GB of 18.0 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
Q8_0	27.5 GB	28.6 GB	—	~2.0	−0.1% ppl	10.6 GB over
Q6_K	21.2 GB	22.3 GB	—	~2.6	−0.4% ppl	4.3 GB over
Q5_K_M	18.4 GB	19.5 GB	—	~3.0	−0.8% ppl	1.5 GB over
Q4_K_M	15.6 GB	16.7 GB	28K	3.6	−1.9% ppl	Recommended
Q3_K_M	12.7 GB	13.8 GB	75K	4.4	−5.4% ppl	Long context
Q2_K	10.8 GB	11.9 GB	104K	5.2	−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 16 of its 64 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

Ollama llama.cpp LM Studio MLX

terminal

$ pip install mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen3.6-27B-4bit \
    --max-tokens 512 --prompt "Hello"

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

01Download is 15.6 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 18.0 GB of 24 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.

03Only 1.3 GB is spare, so a long system prompt can still push it over. Its real ceiling here is 28K context.

See all models for this rig Compare with another model