Runs

Most recent benchmark runs across all devices and models.


Gemopus-4-E4B-it-Preview-Q6_K.gguf Source	M4 Max	llama.cppb8680	63.4 tok/s	1,123.8 tok/s	6.66 GB 19%	Pending
gemma-4-E4B-it.Q8_0.gguf Source	M4 Max	llama.cppb8680	54.4 tok/s	1,180.2 tok/s	8.25 GB 23%	Pending
Gemopus-4-E4B-it-Preview-BF16.gguf Source	M4 Max	llama.cppb8680	33.5 tok/s	1,161.9 tok/s	11.13 GB 31%	Pending
Qwen3.5-9B Source	M4 Max	llama.cppb8680	35.6 tok/s	698.2 tok/s	14.60 GB 41%	Pending
Qwopus3.5-27B-v3 Source	M4 Max	llama.cppb8680	17.3 tok/s	191.3 tok/s	19.51 GB 54%	Verified
Qwopus3.5-27B-v3 Source	M4 Max	llama.cppb8680	14.0 tok/s	189.4 tok/s	22.20 GB 62%	Verified
Qwen3.5-9B Source	M4 Max	mlx_lm0.31.2	60.9 tok/s	707.4 tok/s	8.11 GB 23%	Verified
Qwen3.5-9B Source	M4 Max	llama.cppb8680	35.9 tok/s	696.6 tok/s	9.20 GB 26%	Verified
Qwen3.5-9B Source	M4 Max	llama.cppb8680	21.3 tok/s	707.1 tok/s	17.40 GB 48%	Verified
Gemma 4 E2B IT Source	A18 Pro	llama.cppb8680	13.1 tok/s	194.9 tok/s	0.26 GB 3%	Verified


Gemopus-4-E4B-it-Preview-Q6_K.gguf Source	63.4 tok/s	Pending
gemma-4-E4B-it.Q8_0.gguf Source	54.4 tok/s	Pending
Gemopus-4-E4B-it-Preview-BF16.gguf Source	33.5 tok/s	Pending
Qwen3.5-9B Source	35.6 tok/s	Pending
Qwopus3.5-27B-v3 Source	17.3 tok/s	Verified
Qwopus3.5-27B-v3 Source	14.0 tok/s	Verified
Qwen3.5-9B Source	60.9 tok/s	Verified
Qwen3.5-9B Source	35.9 tok/s	Verified
Qwen3.5-9B Source	21.3 tok/s	Verified
Gemma 4 E2B IT Source	13.1 tok/s	Verified