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

Visual Perception and Reasoning on V* Bench

98.3Attribute Score

Human

Updated 21d ago

Evaluation Results

Method	Links
Human 2026.04		98.3	100	99
TTSP 2026.04		95.7	93.4	94.8
TTSP 2026.04		94.8	92.1	93.7
RTWI 2026.04		94.8	92.1	93.7
RTWI 2026.04		93.9	90.8	92.7
CISC 2026.04		93.9	92.1	93.2
Deepconf 2026.04		93.9	92.1	93.2
CISC 2026.04		93	90.8	92.1
Deepconf 2026.04		93	92.1	92.6
SC 2026.04		93	90.8	92.1
Self-Cer. 2026.04		93	89.5	91.6
SC 2026.04		92.2	90.8	91.6
TRT 2026.04		92.2	90.8	91.6
Base 2026.04		92.2	89.5	91.1
TRT 2026.04		92.2	90.8	91.6
Self-Ref. 2026.04		92.2	90.8	91.6
ZoomEye 2026.03		92.17	86.84	90.05
DeepEyes 2026.04		91.3	88.2	90.1
Base 2026.04		91.3	89.5	90.6
Self-Ref. 2026.04		91.3	89.5	90.6
Self-Cer. 2026.04		91.3	89.5	90.6
DDP 2026.04		89.2	89.5	89.3
SIEVE 2026.03		88.7	86.84	87.96
GRPO 2026.03		86.96	86.84	86.91
DyFo 2026.03		86.09	75	81.68
Zoom-Refine 2026.03		86.09	73.47	81.17
DyFo 2026.03		86.09	78.95	83.25
SCOLAR 2026.05		86.09	81.58	84.29
SCOLAR-7B 2026.05		86.09	80.26	83.77
Vanilla* 2026.03		84.35	80.26	82.72
Deepeyes 2026.05		84.35	81.58	83.25
Thyme 2026.04		83.5	80.3	82.2
Pixel-Reasoner 2026.04		83.5	76.3	80.6
Zoom-Refine 2026.03		83.48	82.89	83.25
Monet 2026.05		83.48	82.89	83.25
GRPO 2026.03		81.74	90.79	85.34
SIEVE 2026.03		81.74	92.11	85.86
ZoomEye 2026.03		81.74	73.68	78.35
LVR 2026.05		81.7	79	80.6
Qwen2.5-VL-7B 2026.05		80	76.31	78.53
Vanilla* 2026.03		74.78	82.89	78.01
GPT-4o 2026.04		72.2	60.5	67.5
GPT-4o 2026.05		72.2	60.5	67.5
GPT-4V 2026.05		61.74	67.11	63.87
GPT-4V 2026.04		51.3	60.5	55
LLaVA-1.5-7B 2026.04		43.5	56.6	48.7
Gemini-1.5-Pro 2026.04		40.9	59.2	48.2
Bard 2026.04		31.3	46.1	37.2
InstructBLIP 2026.04		25.2	47.4	34