MLX-LM on 64GB Apple M4 Max: Qwen3-4B-Instruct-2507 bf16 generation throughput (128 tokens)
52.47 tok/s (generation of 128 tokens, batch 1)
Fields
id
benchmarks/mlx-m4-max-qwen3-4b-bf16-decode-tok-sMetric
decode_tok_s
Value
52.47
Unit
tok/s (generation of 128 tokens, batch 1)
Model
Qwen3-4B-Instruct-2507 (bf16, original Qwen repo)
Methodology
mlx_lm.benchmark --model Qwen/Qwen3-4B-Instruct-2507 -p 2048 -g 128. Batch size 1, 2048 prompt tokens, 128 generated tokens. Hardware: Apple M4 Max 64GB unified memory. Software: mlx 0.29.2.dev20251008+85a8824a8, mlx-lm 0.28.2, macOS 26.1. Peak memory: 9.02 GB. MMLU Pro: 64.05.
Measured by
vendor
Reproducible
yes
notes
Official benchmark published by Apple MLX maintainers in mlx_lm/BENCHMARKS.md. Shows bf16 baseline generation throughput of 52.47 tok/s for Qwen3-4B on 64GB M4 Max, paired with 1780.63 tok/s prefill throughput.
Notes
Official benchmark published by Apple MLX maintainers in mlx_lm/BENCHMARKS.md. Shows bf16 baseline generation throughput of 52.47 tok/s for Qwen3-4B on 64GB M4 Max, paired with 1780.63 tok/s prefill throughput.