Yes — with 1.9 GB to spare

Qwen3.5 4B at Q4_K_M fits your M3 · 8 GB entirely in unified memory at 8K context, at an estimated 21 tokens per second. Past 69K 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 · 2.6 GB Apache 2.0 Released 28 Feb 2026 Vision

The 8 GB coding agent. Q4 lands near 3.4 GB, leaving room for a real context window.

Hardware requirements What hardware do I need to run Qwen3.5 4B? →

At Q4_K_M with an 8K context: a 6 GB GPU (RTX 3060 Laptop) or a 8 GB Mac (M3). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.

The VRAM budget

weights 2.6 GB
Weights 2.6 GB KV cache @ 8K 0.25 GB Runtime overhead 0.6 GB Free 1.9 GB of 5.4 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
F16 8.7 GB 9.5 GB — ~6.4 Reference 4.1 GB over
Q8_0 4.6 GB 5.5 GB 6K ~12 −0.1% ppl 0.1 GB over
Q6_K 3.6 GB 4.4 GB 39K 16 −0.4% ppl Long context
Q5_K_M 3.1 GB 3.9 GB 55K 18 −0.8% ppl Long context
Q4_K_M 2.6 GB 3.5 GB 69K 21 −1.9% ppl Recommended
Q3_K_M 2.1 GB 3.0 GB 85K 26 −5.4% ppl Long context
Q2_K 1.8 GB 2.7 GB 95K 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 8 of its 32 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

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

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

01Download is 2.6 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 5.4 GB of 8 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03There is room to go to 69K context on this card.
See all models for this rig Compare with another model

Related models on an M3 · 8 GB

ModelTotalTok/s est.
Qwen3.5 2B
2.27B · same family
Q4_K_M · 2.0 GB 44 Runs great
Qwen3.5 9B
9.65B · same family
Q4_K_M · 6.3 GB — Won't fit
Qwen3.5 0.8B
0.87B · same family
Q4_K_M · 1.2 GB 114 Runs great
Qwen3.5 122B-A10B
125B MoE · same family
Q4_K_M · 71.1 GB — Won't fit
Gemma 3 4B
4.3B · similar size
Q4_K_M · 3.3 GB 23 Runs great
Gemma 4 E2B
5.1B · similar size
Q4_K_M · 3.5 GB 19 Runs great
Qwen3 4B
4.02B · similar size
Q4_K_M · 4.0 GB 25 Runs great

Every model against the M3 · 8 GB →

Qwen3.5 4B on other hardware

DeviceTotalTok/s est.
GeForce RTX 3060 Laptop
5.3 GB usable · 336 GB/s
3.5 GB 78 Runs great
GeForce RTX 4060 Laptop
7.0 GB usable · 256 GB/s
3.5 GB 59 Runs great
GeForce RTX 3060 12 GB
10.6 GB usable · 360 GB/s
3.5 GB 83 Runs great
M4 · 16 GB
10.7 GB usable · 120 GB/s
3.5 GB 26 Runs great
GeForce RTX 4090
22.4 GB usable · 1008 GB/s
3.5 GB 233 Runs great
Radeon RX 7900 XTX
22.4 GB usable · 960 GB/s
3.5 GB 222 Runs great
M3 Max · 64 GB
48.0 GB usable · 400 GB/s
3.5 GB 85 Runs great

All 114 devices, by tier →