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

On the Apple M3 Ultra, local/qwen3.8-flash-next ranks first for coding in our benchmarks (32.7/100 at ~74.6 tok/s).

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

On the Apple M3 Ultra, local/qwen3.8-flash-next ranks first for agentic workflows in our benchmarks (75.7/100 at ~74.6 tok/s).

Hardware Benchmarks

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

GPU-specific specifications for local LLM planning.

Apple M3 Ultra

Models we recommend for Apple M3 Ultra.

Fastest & Best quality

Unknown

74.7 tok/sQuality 68.7

Highest token generation speed at 74.7 tok/s. Highest completed quality score at 68.7.

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

What is the best local model for coding on the Apple M3 Ultra?: On the Apple M3 Ultra, local/qwen3.8-flash-next ranks first for coding in our benchmarks (32.7/100 at ~74.6 tok/s).
What is the best local model for agentic workflows on the Apple M3 Ultra?: On the Apple M3 Ultra, local/qwen3.8-flash-next ranks first for agentic workflows in our benchmarks (75.7/100 at ~74.6 tok/s).

Completed quality-ranked benchmark results for Apple M3 Ultra.