Yes — with 89.7 GB to spare
Qwen3 8B at Q4_K_M fits your M4 Max · 128 GB entirely in unified memory at 8K context, at an estimated 66 tokens per second. There is room for its full 128K window.
Fully in unified memory
8K context
Q4_K_M · 4.6 GB
Apache 2.0
Released Apr 2025
Apache-2.0 alternative to Llama 3.1 8B, with a switchable thinking mode.
Hardware requirements What hardware do I need to run Qwen3 8B? →At Q4_K_M with an 8K context: a 10 GB GPU (Arc B570) or a 16 GB Mac (M4). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.
The VRAM budget
weights 4.6 GB
Weights 4.6 GB
KV cache @ 8K 1.13 GB
Runtime overhead 0.6 GB
Free 89.7 GB of 96.0 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 15.3 GB | 17.0 GB | 128K | 20 | Reference | Long context |
| Q8_0 | 8.1 GB | 9.8 GB | 128K | 38 | −0.1% ppl | Long context |
| Q6_K | 6.3 GB | 8.0 GB | 128K | 49 | −0.4% ppl | Long context |
| Q5_K_M | 5.4 GB | 7.1 GB | 128K | 56 | −0.8% ppl | Long context |
| Q4_K_M | 4.6 GB | 6.3 GB | 128K | 66 | −1.9% ppl | Recommended |
| Q3_K_M | 3.7 GB | 5.5 GB | 128K | 82 | −5.4% ppl | Long context |
| Q2_K | 3.2 GB | 4.9 GB | 128K | 96 | −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.
How to run it
$ pip install mlx-lm $ mlx_lm.generate --model mlx-community/Qwen3-8B-4bit \ --max-tokens 512 --prompt "Hello"
Apple's own array framework. The fastest path on Apple Silicon. More on MLX.
01Download is 4.6 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 96.0 GB of 128 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03There is room to go to the model's full 128K context on this card.
Related models on an M4 Max · 128 GB
| Model | Total | Tok/s est. | |
|---|---|---|---|
| Qwen3 14B 14.8B · same family |
Q4_K_M · 10.2 GB | 37 | Runs great |
| Qwen3 4B 4.02B · same family |
Q4_K_M · 4.0 GB | 135 | Runs great |
| Qwen3 30B-A3B 30.5B MoE · same family |
Q4_K_M · 18.5 GB | 77 | Runs great |
| Qwen3 Coder 30B-A3B 30.5B MoE · same family |
Q4_K_M · 18.5 GB | 77 | Runs great |
| Fara 7B 8.29B · similar size |
Q4_K_M · 5.7 GB | 65 | Runs great |
| Llama 3.1 8B Instruct 8.03B · similar size |
Q4_K_M · 6.1 GB | 68 | Runs great |
| Gemma 4 E4B 8.0B · similar size |
Q4_K_M · 5.2 GB | 68 | Runs great |
Qwen3 8B on other hardware
| Device | Total | Tok/s est. | |
|---|---|---|---|
| GeForce RTX 4060 Laptop 7.0 GB usable · 256 GB/s |
6.3 GB | 34 | Tight fit |
| Arc B570 8.8 GB usable · 380 GB/s |
6.3 GB | 50 | Runs great |
| GeForce RTX 3060 12 GB 10.6 GB usable · 360 GB/s |
6.3 GB | 47 | Runs great |
| M4 · 16 GB 10.7 GB usable · 120 GB/s |
6.3 GB | 15 | Runs great |
| GeForce RTX 4090 22.4 GB usable · 1008 GB/s |
6.3 GB | 133 | Runs great |
| Radeon RX 7900 XTX 22.4 GB usable · 960 GB/s |
6.3 GB | 126 | Runs great |
| M3 Max · 64 GB 48.0 GB usable · 400 GB/s |
6.3 GB | 49 | Runs great |