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

High-resolution perception on V*

89.53Overall Score

TikArt

Updated 1d ago

Evaluation Results

Method
TikArt 2026.02	89.53	91.3	86.84
Gemini-3-Flash 2026.04	86.4	-	-
TikArt w/o Observation 2026.02	86.39	89.57	81.58
DeepEyes 2026.04	85.6	-	-
DeepEyes-7B 2026.05	85.6	-	-
MiMo-VL-7B +PC 2026.03	85.3	-	-
DeepLatent-RL-7B* 2026.05	85.3	-	-
Qwen3-VL-Instruct 2026.02	84.82	84.35	85.53
DeepLatent-RL-7B 2026.05	84.8	-	-
Qwen3-VL-235B-A22B-Instruct 2026.02	84.66	88.5	78.95
Pixel-Reasoner 2026.02	84.3	-	-
SkiLa 2026.04	84.3	-	-
Pixel-Reasoner-7B 2026.05	84.3	-	-
DyFo-7B 2026.05	84.3	-	-
DeepLatent-SFT-7B 2026.05	84.3	-	-
LLaVA-OneVision-1.5 2026.02	84.29	83.48	85.53
HyLaR-7B 2026.04	83.77	82.61	85.53
Monet-7B 2026.05	83.3	-	-
Monet-7B 2026.02	83.25	83.48	82.89
Monet 2026.04	83.25	83.48	82.89
TikArt w/o Segment Action 2026.02	83.24	82.6	84.21
Deepeyes 2026.02	83.03	84.35	81.58
Thyme 2026.02	82.2	-	-
Thyme 2026.04	82.2	83.5	80.3
Thyme-7B 2026.05	82.2	-	-
DeepEyesV2 2026.04	81.8	-	-
LVR-7B 2026.05	81.7	-	-
InternVL3-8B 2026.05	81.2	-	-
HyLaR-SFT 2026.04	80.63	81.73	78.95
LVR 2026.02	80.6	81.7	79
MiMo-VL-7B 2026.03	80.6	-	-
LVR 2026.04	80.6	81.7	79
TikArt w/o Zoom Action 2026.02	80.34	76.32	86.42
Gemini-2.5-flash 2026.02	80.1	84.35	73.68
Monet 2026.04	80.1	81.73	77.63
ZoomEye 2026.04	79.85	80.52	78.82
Qwen2.5-VL-7B +PC 2026.03	79.7	-	-
Gemini-2.5-pro 2026.02	79.06	84.35	71.05
SkiLa 2026.04	78.53	-	-
CoVT-7B 2026.05	78.5	-	-
Qwen2.5-VL-7B 2026.05	76.5	-	-
GPT-5 2026.02	76.44	74.78	78.95
Qwen2.5-VL-7B 2026.04	76.44	77.39	75
Qwen2.5-VL-7B 2026.03	76.4	-	-
Qwen2.5-VL-3B 2026.03	75.4	-	-
LLaVA-OneVision-7B 2026.05	75.4	-	-
TikArt w/o GRPO 2026.02	75.39	78.26	71.05
Qwen3-VL-Instruct 2026.02	73.82	73.04	73.68
Qwen2.5-VL-3B +PC 2026.03	73.8	-	-
Qwen3-VL-Thinking 2026.02	72.77	71.3	75
LLaVA-OneVision-7B 2026.04	71.25	73.48	67.89
GPT-4o 2026.02	70.68	71.3	69.74
InternVL3-8B 2026.04	70.2	67.8	73.7
GPT-4o 2026.04	67.5	72.2	60.5
GPT-5-nano 2026.02	63.87	58.26	72.37