Yes — with 16.0 GB to spare

Qwen3.5 2B at Q4_K_M fits your M3 · 24 GB entirely in unified memory at 8K context, at an estimated 44 tokens per second. There is room for its full 256K window.

Fully in unified memory 8K context Q4_K_M · 1.3 GB Apache 2.0 Released 28 Feb 2026 Vision

Phone-class, and multimodal. Replaces Llama 3.2 3B as the "it runs on anything" answer.

What hardware do I need for Qwen3.5 2B? →

The VRAM budget

weights 1.3 GB

Weights 1.3 GB KV cache @ 8K 0.09 GB Runtime overhead 0.6 GB Free 16.0 GB of 18.0 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
F16	4.2 GB	4.9 GB	256K	13	Reference	Long context
Q8_0	2.2 GB	2.9 GB	256K	25	−0.1% ppl	Long context
Q6_K	1.7 GB	2.4 GB	256K	32	−0.4% ppl	Long context
Q5_K_M	1.5 GB	2.2 GB	256K	37	−0.8% ppl	Long context
Q4_K_M	1.3 GB	2.0 GB	256K	44	−1.9% ppl	Recommended
Q3_K_M	1.0 GB	1.7 GB	256K	54	−5.4% ppl	Long context
Q2_K	0.9 GB	1.6 GB	256K	63	−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 6 of its 24 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/Qwen3.5-2B-4bit \
    --max-tokens 512 --prompt "Hello"

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

01Download is 1.3 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.

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