Share your thoughts, 1 month free Claude Pro on usSee more

Image-to-Code Generation on STEM2Code-Eval (1k samples)

68.97Image Score

CodePercept-32B-R1

Updated 1mo ago

Evaluation Results

Method
CodePercept-32B-R1 2026.03	68.97	62.53	0.959	75.8
CodePercept-32B-R1 2026.03	68.97	62.53	95.9	65.75
Gemini2.5-Pro-Thinking 2026.03	68.89	75.41	0.917	78.67
Gemini2.5-Pro 2026.03	68.89	75.41	91.7	72.15
GPT5-Thinking 2026.03	64.97	64.98	0.966	75.52
GPT5 2026.03	64.97	64.98	96.6	64.98
CodePercept-32B-S1 2026.03	61.14	56.99	0.93	70.38
CodePercept-32B-S1 2026.03	61.14	56.99	93	59.07
Gemini2.5-Flash-Thinking 2026.03	57.25	60.87	0.852	67.77
Gemini2.5-Flash 2026.03	57.25	60.87	85.2	59.06
Claude-Opus 4.1-Thinking 2026.03	55.9	56.19	0.971	69.73
Claude-Opus 4.1 2026.03	55.9	56.19	97.1	56.05
CodePercept-8B-R1 2026.03	50.25	47.04	0.934	63.56
CodePercept-8B-R1 2026.03	50.25	47.04	93.4	48.65
CodePercept-4B-R1 2026.03	47.17	45.86	0.913	61.44
CodePercept-4B-R1 2026.03	47.17	45.86	91.3	46.52
Qwen3-VL-Plus-Instruct 2026.03	45.94	40.4	0.9	58.78
Qwen3-VL-Plus 2026.03	45.94	40.4	90	43.17
Qwen3-VL-Plus-Thinking 2026.03	45.59	40.61	0.892	58.47
Qwen3-VL-Plus 2026.03	45.59	40.61	89.2	43.1
CodePercept-8B-S1 2026.03	44.53	46.78	0.876	59.64
CodePercept-8B-S1 2026.03	44.53	46.78	87.6	45.66
Seed1.6-Vision-Thinking 2026.03	42.03	40.74	0.947	59.15
Seed1.6-Vision 2026.03	42.03	40.74	94.7	41.39
CodePercept-4B-S1 2026.03	38.13	43.43	0.807	54.09
CodePercept-4B-S1 2026.03	38.13	43.43	80.7	40.78
Qwen3-VL-30A3B-Thinking 2026.03	37.47	35.53	0.871	53.37
Qwen3-VL-30A3B 2026.03	37.47	35.53	87.1	36.5
Qwen3-VL-32B-Instruct 2026.03	36.85	39.98	0.818	52.88
Qwen3-VL-32B 2026.03	36.85	39.98	81.8	38.42
Qwen3-VL-30A3B-Instruct 2026.03	33.05	31.04	0.875	50.53
Qwen3-VL-30A3B 2026.03	33.05	31.04	87.5	32.05
Qwen2.5-VL-72B-Instruct 2026.03	32.82	25.83	0.863	48.32
Qwen2.5-VL-72B 2026.03	32.82	25.83	86.3	29.33
Seed1.6-Vision-nothinking 2026.03	31.22	38.56	0.855	51.76
Seed1.6-Vision 2026.03	31.22	38.56	85.5	34.89
Qwen3-VL-8B-Thinking 2026.03	29.82	41.71	0.789	50.14
Qwen3-VL-8B 2026.03	29.82	41.71	78.9	35.77
Qwen3-VL-8B-Instruct 2026.03	28.59	28.23	0.853	47.37
Qwen3-VL-8B 2026.03	28.59	28.23	85.3	28.41
Qwen3-VL-4B-Thinking 2026.03	25.38	34.53	0.757	45.2
Qwen3-VL-4B 2026.03	25.38	34.53	75.7	29.96
Qwen3-VL-4B-Instruct 2026.03	24.55	26.42	0.794	43.46
Qwen3-VL-4B 2026.03	24.55	26.42	79.4	25.49
GLM-4.1V-9B 2026.03	21.19	26.51	0.72	39.9
GLM-4.1V-9B 2026.03	21.19	26.51	72	23.85
KeyeVL1.5-8B 2026.03	20.33	22.47	0.734	38.73
KeyeVL1.5-8B 2026.03	20.33	22.47	73.4	21.4
MiMo-VL-7B-RL 2026.03	14.54	22.41	0.603	32.42
MiMo-VL-7B-RL 2026.03	14.54	22.41	60.3	18.48
MiniCPM-V-4.5 2026.03	13.91	23.69	0.508	29.47
MiniCPM-V-4.5 2026.03	13.91	23.69	50.8	18.8
InternVL3.5-8B 2026.03	13.5	18.16	0.565	29.38
InternVL3.5-8B 2026.03	13.5	18.16	56.5	15.83
Ovis2.5-9B 2026.03	9.76	11.26	0.894	36.81
Ovis2.5-9B 2026.03	9.76	11.26	89.4	10.51
Intern-S1-8B 2026.03	6.02	15.87	0.266	16.16
Intern-S1-8B 2026.03	6.02	15.87	26.6	10.95