M3 Max — benchmark record

M3 Max local LLM benchmarks across 36 GB, 96 GB RAM tiers on Apple Silicon Mac. 4 published rows across 4 models with explicit evidence state and RAM-tier comparison. Peak published speed is 28.0 tok/s.

4Benchmark rows

4Models tested

2RAM configurations

28.0Fastest avg tok/s

RAM configurations

Each configuration differs only in unified memory. More RAM = larger models fit. Throughput is similar across RAM tiers at the same model size.

36 GBM3 Max (36 GB)2 rows · 2 models28.0 tok/s peak 96 GBM3 Max (96 GB)2 rows · 2 models28.0 tok/s peak

All benchmark rows — M3 Max

Sorted by avg tok/s descending. Click source badge to see original measurement.

Chip (RAM)	Model	Quant	RAM req.	Avg tok/s	Prompt tok/s	Runtime	Source
M3 Max (96 GB)	Gemma 3 27B	Q4_K - Medium	—	28.0 tok/s	—	MLX	ref
M3 Max (36 GB)	Qwen 3 30B-A3B	Q4_K - Medium	—	28.0 tok/s	—	Ollama	ref
M3 Max (96 GB)	Qwen3.5-35B-A3B	Q4_K - Medium	—	22.0 tok/s	—	Ollama	ref
M3 Max (36 GB)	DeepSeek R1 Distill Qwen 32B	Q4_K - Medium	—	14.0 tok/s	—	Ollama	ref

Models tested on M3 Max

DeepSeek R1 Distill Qwen 32B Gemma 3 27B Qwen 3 30B-A3B Qwen3.5-35B-A3B

Data

benchmarks.json — full dataset · chips.json — chip summaries · benchmarks.csv — CSV export

Data from in-house lab measurements plus community-published benchmarks. See all chip families →