Yes — with 181.8 GB to spare
Qwen3 14B at Q4_K_M fits your M3 Ultra · 256 GB entirely in unified memory at 8K context, at an estimated 54 tokens per second. There is room for its full 128K window.
Fully in unified memory
8K context
Q4_K_M · 8.3 GB
Apache 2.0
Released Apr 2025
The largest Qwen3 that fits a 12 GB card at Q4 with room for context.
Hardware requirements What hardware do I need to run Qwen3 14B? →At Q4_K_M with an 8K context: a 16 GB GPU (RTX 5080) or a 24 GB Mac (M4 Pro). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.
The VRAM budget
weights 8.3 GB
Weights 8.3 GB
KV cache @ 8K 1.25 GB
Runtime overhead 0.6 GB
Free 181.8 GB of 192.0 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 27.6 GB | 29.4 GB | 128K | 16 | Reference | Long context |
| Q8_0 | 14.6 GB | 16.5 GB | 128K | 31 | −0.1% ppl | Long context |
| Q6_K | 11.3 GB | 13.2 GB | 128K | 40 | −0.4% ppl | Long context |
| Q5_K_M | 9.8 GB | 11.6 GB | 128K | 46 | −0.8% ppl | Long context |
| Q4_K_M | 8.3 GB | 10.2 GB | 128K | 54 | −1.9% ppl | Recommended |
| Q3_K_M | 6.7 GB | 8.6 GB | 128K | 66 | −5.4% ppl | Long context |
| Q2_K | 5.8 GB | 7.6 GB | 128K | 77 | −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/Qwen3-14B-4bit \ --max-tokens 512 --prompt "Hello"
Apple's own array framework. The fastest path on Apple Silicon. More on MLX.
01Download is 8.3 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 192.0 GB of 256 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 M3 Ultra · 256 GB
| Model | Total | Tok/s est. | |
|---|---|---|---|
| Qwen3 8B 8.19B · same family |
Q4_K_M · 6.3 GB | 97 | Runs great |
| Qwen3 30B-A3B 30.5B MoE · same family |
Q4_K_M · 18.5 GB | 112 | Runs great |
| Qwen3 Coder 30B-A3B 30.5B MoE · same family |
Q4_K_M · 18.5 GB | 112 | Runs great |
| Qwen3 32B 32.8B · same family |
Q4_K_M · 21.0 GB | 24 | Runs great |
| DeepSeek-R1-Distill-Qwen 14B 14.8B · similar size |
Q4_K_M · 10.4 GB | 54 | Runs great |
| Qwen2.5-Coder 14B 14.8B · similar size |
Q4_K_M · 10.4 GB | 54 | Runs great |
| Phi-4 14B 14.7B · similar size |
Q4_K_M · 10.4 GB | 54 | Runs great |
Qwen3 14B on other hardware
| Device | Total | Tok/s est. | |
|---|---|---|---|
| GeForce RTX 4060 Laptop 7.0 GB usable · 256 GB/s |
10.2 GB | ~8.3 | CPU offload |
| GeForce RTX 3060 12 GB 10.6 GB usable · 360 GB/s |
10.2 GB | 26 | Tight fit |
| GeForce RTX 5080 14.4 GB usable · 960 GB/s |
10.2 GB | 70 | Runs great |
| M4 · 16 GB 10.7 GB usable · 120 GB/s |
10.2 GB | 8.1 | Tight fit |
| GeForce RTX 4090 22.4 GB usable · 1008 GB/s |
10.2 GB | 73 | Runs great |
| Radeon RX 7900 XTX 22.4 GB usable · 960 GB/s |
10.2 GB | 70 | Runs great |
| M4 Pro · 24 GB 18.0 GB usable · 273 GB/s |
10.2 GB | 18 | Runs great |
| M3 Max · 64 GB 48.0 GB usable · 400 GB/s |
10.2 GB | 27 | Runs great |