text model · Qwen3 · macOS
Can I run Qwen3 235B A22B on Apple M3 Ultra (256GB)?
Yes. Qwen3 235B A22B runs on Apple M3 Ultra (256GB) at Q4_K_M (~136.9 GB of ~192 GB usable).
Runs at Q4_K_M using ~136.9 GB of ~192 GB usable.
That figure is at a 4k context and moves about ±15% as context length changes. Apple M3 Ultra (256GB) leaves ~55.1 GB of headroom.
- Q4_K_M needed
- ~136.9 GB
- Usable on device
- ~192 GB
- Device memory
- 256 GB
- Best quant
- Q4_K_M
Which quant fits
Run it
Pick your tool. All 3 load the same Q4_K_M weights.
ollama run qwen3:235b llama-cli -hf unsloth/Qwen3-235B-A22B-GGUF:Q4_K_M lms get unsloth/Qwen3-235B-A22B-GGUF - Parameters
- 235B (MoE, 22B active)
- Q4_K_M size
- 132.39 GB
- Q8_0 size
- 232.77 GB
- Context
- 128k
- Ollama tag
- qwen3:235b
- Memory
- 256 GB unified
- Usable for weights
- ~192 GB
- Power draw
- ~270 W
- Best runtime
- MLX direct / Ollama (MLX backend)
You could also run
FAQ
Can Apple M3 Ultra (256GB) run Qwen3 235B A22B?
Yes. Qwen3 235B A22B runs on Apple M3 Ultra (256GB) at Q4_K_M (~136.9 GB of ~192 GB usable).
How much memory does Qwen3 235B A22B need?
Apple M3 Ultra (256GB) has room to spare. At Q4_K_M the weights are ~132.39 GB; with KV cache and runtime overhead, budget ~136.9 GB at a 4k context. It is a Mixture-of-Experts model (235B total / 22B active), so all experts must stay in memory; memory tracks total params, not active params.
What is the best tool to run Qwen3 235B A22B on macOS?
LM Studio for a simple setup; mlx-lm for the most speed. vLLM is NOT a Mac tool, it is a CUDA/Linux serving engine. Unified memory is not a fixed VRAM slice; ~70% is usable for weights.
Embed this
A live badge for your model card or README, updated as the data is. [](https://localmodel.run/can-i-run/qwen3-235b-a22b/apple-m3-ultra-256gb) Sources
Weights are measured from GGUF files; KV cache and overhead are computed, so totals can vary ~15% with context and runtime. Any tok/s is a bandwidth estimate. See methodology.