RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Qwen3 0.6B run on the Mac M3 Pro 36GB?

Yes. At 8K context the best fit is F16, using about 2.7 GB of the 24.1 GB usable on the Mac M3 Pro 36GB. Estimated speed ceiling: ~125 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •1.4 GB Fits1.8 GB Fits4.5 GB Fits506
3-bit (Q3_K_M) •1.5 GB Fits1.9 GB Fits4.5 GB Fits432
4-bit (Q4_K_M) •1.5 GB Fits1.9 GB Fits4.6 GB Fits378
5-bit (Q5_K_M) •1.5 GB Fits2.0 GB Fits4.6 GB Fits338
6-bit (Q6_K) •1.6 GB Fits2.0 GB Fits4.7 GB Fits303
8-bit (Q8_0) •1.7 GB Fits2.2 GB Fits4.8 GB Fits235
16-bit (F16/BF16) •2.2 GB Fits2.7 GB Fits5.3 GB Fits125

• real GGUF file size from unsloth/Qwen3-0.6B-GGUF. Speed is a bandwidth ceiling (150 GB/s), not a benchmark. Methodology.

More

Qwen3 0.6B on every GPU and Mac · Every model that runs on the Mac M3 Pro 36GB · Apple Silicon guide · Try other settings in the calculator

Data: Hugging Face Hub (Qwen/Qwen3-0.6B), Apple specs. Data as of 2026-11-17.