Qwen
Qwen
Qwen/Qwen3-4B-Instruct-2507

Runs

View all benchmark runs for this model family.

	Quant						Actions

	Quant		Actions

	Quant						Actions
M5 Pro	4bit	mlx_lm0.31.2	82.9 tok/s	2,359.8 tok/s	6.12 GB 13%
M3	q8_0	llama.cppb8480	11.2 tok/s	234.0 tok/s	0.71 GB 3%
M1 Max	q8_0	llama.cppb8240	30.4 tok/s	495.6 tok/s	5.02 GB 8%
M1 Max	4bit	mlx_lm0.31.0	39.9 tok/s	290.2 tok/s	4.08 GB 6%

	Quant		Actions
M5 Pro	4bit	82.9 tok/s
M3	q8_0	11.2 tok/s
M1 Max	q8_0	30.4 tok/s
M1 Max	4bit	39.9 tok/s