M1 Max — benchmark record

M1 Max local LLM benchmarks across 64 GB RAM tiers on Apple Silicon Mac. 10 published rows across 5 models with explicit evidence state and RAM-tier comparison. Peak published speed is 58.5 tok/s.

10Benchmark rows

5Models tested

1RAM configurations

58.5Fastest 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.

64 GBM1 Max (64 GB)10 rows · 5 models58.5 tok/s peak

All benchmark rows — M1 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
M1 Max (64 GB)	Qwen3-Coder-30B-A3B	IQ4_XS	16.1 GB	58.5 tok/s	132.1 tok/s	llama.cpp	ref
M1 Max (64 GB)	Qwen3.5-35B-A3B	4bit	—	57.6 tok/s	431.0 tok/s	MLX	ref
M1 Max (64 GB)	Qwen3.5-35B-A3B	4bit	—	57.0 tok/s	—	LM Studio (MLX)	ref
M1 Max (64 GB)	Nemotron-3-Nano-30B-A3B	Q4_K_XL	22.0 GB	43.7 tok/s	136.9 tok/s	llama.cpp	ref
M1 Max (64 GB)	GLM-4.7-Flash	Q4_K_XL	17.0 GB	36.8 tok/s	99.4 tok/s	llama.cpp	ref
M1 Max (64 GB)	Qwen3.5-35B-A3B	Q4_K - Medium	—	29.0 tok/s	—	LM Studio (llama.cpp)	ref
M1 Max (64 GB)	Qwen3.5-27B	4bit	—	15.0 tok/s	67.0 tok/s	MLX	ref
M1 Max (64 GB)	Qwen3.5-27B	Q6_K	—	12.0 tok/s	—	llama.cpp	ref
M1 Max (64 GB)	Qwen3.5-27B	Q4_K - Medium	—	11.5 tok/s	—	llama.cpp	ref
M1 Max (64 GB)	Qwen3.5-27B	Q8_0	—	10.5 tok/s	—	llama.cpp	ref

Models tested on M1 Max

GLM-4.7-Flash Nemotron-3-Nano-30B-A3B Qwen3-Coder-30B-A3B Qwen3.5-27B 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 →