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

Image Reasoning on WeMath

71.3Accuracy

Gemini-2.5-Pro

Updated 19d ago

Evaluation Results

Method	Links
Gemini-2.5-Pro 2026.05		71.3
Claude-4-Sonnet 2026.05		63
OPD V→T 2026.04		60.95
MOPD 2026.04		59.9
CoPD 2026.04		59.9
CoPD 2026.04		59.81
Image-Expert 2026.04		59.14
Image-Expert 2026.04		59.14
Mixed RLVR 2026.04		59.05
OPD T→V 2026.04		58.86
Mixed RLVR 2026.04		57.43
Text-Expert 2026.04		55.24
Text-Expert 2026.04		55.24
Video-Expert 2026.04		53.43
Base 2026.04		52.38
Base 2026.04		52.38
GPT-4o 2026.05		50.6
Gemini-2.0-Flash 2026.05		47.4
ATLASLA-GRPO 2026.05		45
VTS-V 2026.05		42.8
LVR 2026.05		41.2
ATLASGRPO 2026.05		40.3
MCOT 2026.05		39.6
Bagel 2026.05		39.4
CoVT 2026.05		38.1
Monet 2026.05		36.9
Qwen2.5-VL 2026.05		36.2
V-Thinker 2026.05		32.5
Gemma-3-27B 2026.05		31.7
ATLASSFT 2026.05		28.9
Visual CoT 2026.05		28.6
Anole 2026.05		24.7
LLaVA-OneVision-7B 2026.05		23.1
MiniGPT-v2 2026.05		11