Runs | whatcani.run

Runs

Most recent benchmark runs across all devices and models.

								Actions
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source

			Actions
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source
Source

								Actions
Gemma 4 E2B IT Source	M3 Pro	llama.cppb9430	62.5 tok/s	1,071.9 tok/s	3.95 GB 11%	Pending
Qwen3.6-27B Source	M5 Max	mlx_lm0.31.3	14.8 tok/s	509.7 tok/s	36.00 GB 28%	Pending
Qwen3.5-4B Source	M4 Pro	llama.cppb9260	53.3 tok/s	707.1 tok/s	0.72 GB 3%	Pending
LFM2.5-350M Source	M4 Pro	llama.cppb9260	370.4 tok/s	8,613.4 tok/s	0.50 GB 2%	Pending
Qwen3.5-2B Source	M4	llama.cppb9260	52.4 tok/s	860.0 tok/s	0.85 GB 4%	Verified
Llama 3.2 3B Instruct Source	M5	mlx_lm0.31.3	46.5 tok/s	1,957.2 tok/s	3.48 GB 22%	Verified
Qwen3.5-4B Source	M4	llama.cppb9260	25.9 tok/s	321.1 tok/s	1.10 GB 5%	Verified
Gemma 4 31B Source	M5 Pro	mlx_lm0.31.3	14.6 tok/s	364.6 tok/s	22.00 GB 34%	Verified
Qwen3.6-27B Source	M1 Max	llama.cppb8180	8.3 tok/s	117.8 tok/s	4.06 GB 13%	Verified
Gemma 4 E4B IT Source	GeForce RTX 2080 Ti	llama.cppb9110	93.3 tok/s	3,378.0 tok/s	5.08 GB 16%	Verified

			Actions
Gemma 4 E2B IT Source	62.5 tok/s	Pending
Qwen3.6-27B Source	14.8 tok/s	Pending
Qwen3.5-4B Source	53.3 tok/s	Pending
LFM2.5-350M Source	370.4 tok/s	Pending
Qwen3.5-2B Source	52.4 tok/s	Verified
Llama 3.2 3B Instruct Source	46.5 tok/s	Verified
Qwen3.5-4B Source	25.9 tok/s	Verified
Gemma 4 31B Source	14.6 tok/s	Verified
Qwen3.6-27B Source	8.3 tok/s	Verified
Gemma 4 E4B IT Source	93.3 tok/s	Verified