Can I run it?
Can I run Qwen3.8 35B A3B Distill on Apple M4 Pro?
Tight fit
Tight fit fully in unified memory on your Apple M4 Pro at Q3_K_M, 8K context, about 90–123 tok/s. Weights 16.5 GB plus 0.19 GB of KV cache — 17.1 GB of 18.0 GB usable. Up to 32K context still fits.
Best quantQ3_K_M
Working context8K
Max that fits32K
Speed90–123 tok/s
Weights16.5 GB
KV cache0.19 GB
Total17.1 GB
Usable18.0 GB
Qwen3.8 35B A3B Distill details · Apple M4 Pro page · How this estimate is made
Run commands
llama.cpp
llama-server -hf empero-ai/Qwen3.8-35B-A3B-Distill-GGUF:Q3_K_M -c 8192 -ngl 40
Ollama
ollama run qwen3.8
LM Studio
lms get empero-ai/Qwen3.8-35B-A3B-Distill-GGUF # Then load quant Q3_K_M
Assumptions
- Estimate, not a benchmark. Decode speed uses a memory-bandwidth model at 55–75% of peak.
- KV cache is FP16. Engine overhead assumes llama.cpp (0.3 GB + 1% of the weights). Ollama is closer to 0.9 GB + 3%, LM Studio closer to 1 GB + 3%.
- Q3_K_M size is the real GGUF file size from Hugging Face.
- Unified memory uses 75% for the GPU by default, and always leaves about 5 GB for the OS.
- Speed is quoted at 8K context, not at the maximum context that still fits.
Larger models that also fit
- Can I run Llama 3.3 70B Instruct on Apple M4 Pro? — Tight fit, UD-IQ1_S, 8.7–12 tok/s
- Can I run DeepSeek R1 Distill Llama 70B on Apple M4 Pro? — Tight fit, UD-IQ1_S, 8.7–12 tok/s
- Can I run Hermes 4 70B on Apple M4 Pro? — Tight fit, UD-IQ1_S, 8.7–12 tok/s
Updated 2026-10-17. Profile on this page is the device default.