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

Agentic Multimodal Tool-use on RealX-Bench

70Average Score

Human

Updated 1mo ago

Evaluation Results

Method	Links
Human 2025.11		70	69.5	63.5	62.1	51.4
Gemini 2.5 Pro 2025.11		46	41.5	33.7	43.6	27.8
Gemini 2.5 Pro 2025.11		45	37.8	33.2	43.1	25
Gemini 2.5 Pro 2025.11		41.7	39	28.7	38.4	23.6
o3 2025.11		41.3	37.8	28.1	40.3	22.2
o3 2025.11		41	34.8	28.7	37.9	19.4
Gemini 2.5 Pro 2025.11		39.3	34.8	24.2	36	16.7
o3 2025.11		39.3	38.4	25.3	37	20.8
GPT4o 2025.11		38.7	30.5	27.5	36.5	15.3
GPT4o 2025.11		36.3	29.3	25.8	36.5	16.7
o3 2025.11		35	31.7	23	30.8	11.1
GPT4o 2025.11		32.3	29.9	22.5	29.4	16.7
GPT4o 2025.11		32	29.3	23	29.4	16.7
Qwen2.5-VL-32B 2025.11		32	27.4	29.2	31.8	23.6
Qwen2.5-VL-72B 2025.11		31	35.4	25.8	25.6	23.6
Qwen2.5-VL-32B 2025.11		30.7	27.4	23	26.1	15.3
DeepEyesV2 2025.11		28.3	19.5	22.5	28.9	18.1
Qwen2.5-VL-72B 2025.11		28	28.7	20.2	20.4	15.3
Qwen2.5-VL-72B 2025.11		26.3	28.7	19.7	20.4	16.7
Qwen2.5-VL-32B 2025.11		25.7	25.6	20.2	19	16.7
Qwen2.5-VL-72B 2025.11		25.3	23.1	17.4	17.5	9.7
Qwen2.5-VL-32B 2025.11		25	21.3	19.7	19.9	12.5
Qwen2.5-VL-7B 2025.11		22.3	17.1	16.3	19.9	9.7
Qwen2.5-VL-7B 2025.11		21.7	17.7	15.7	18.5	7.6
Thyme 2025.11		21	18.3	14.6	12.8	4.2
Qwen2.5-VL-7B 2025.11		19.7	16.6	14.4	15.9	8.3
DeepEyes 2025.11		19	19.5	14.6	12.8	9.7
Qwen2.5-VL-7B 2025.11		17	15.9	13.5	12.3	6.9