What is the best local model for coding on the Apple M3 Pro?

On the Apple M3 Pro, qwen3.5:9b-q4_K_M ranks first for coding in our benchmarks (44.2/100 at ~22.5 tok/s).

What is the best local model for agentic workflows on the Apple M3 Pro?

On the Apple M3 Pro, qwen3.5:9b-q4_K_M ranks first for agentic workflows in our benchmarks (77.2/100 at ~22.5 tok/s).

Hardware Benchmarks

Compare Apple M3 Pro benchmark results and see which models actually earn the best completed quality scores.

GPU-specific specifications for local LLM planning.

Apple M3 Pro

Models we recommend for Apple M3 Pro.

Fastest

4.2B

36.1 tok/s

Highest token generation speed at 36.1 tok/s.

Best quality & Largest context

9.0B

22.5 tok/sQuality 70.9Context 65,536 tokens

Highest completed quality score at 70.9. Largest context that fits in 36GB VRAM, at 65,536 tokens.

Which models actually run best on the Apple M3 Pro, by task, from community benchmark data.

What is the best local model for coding on the Apple M3 Pro?: On the Apple M3 Pro, qwen3.5:9b-q4_K_M ranks first for coding in our benchmarks (44.2/100 at ~22.5 tok/s).
What is the best local model for agentic workflows on the Apple M3 Pro?: On the Apple M3 Pro, qwen3.5:9b-q4_K_M ranks first for agentic workflows in our benchmarks (77.2/100 at ~22.5 tok/s).

Completed quality-ranked benchmark results for Apple M3 Pro.