List your company

Geometry problem solving on MathVista GPS

95.7AccuracyDreamPRM (o4-mini)
13.22834.63956.0577.461May 15, 2025May 17, 2025May 19, 2025May 22, 2025May 24, 2025May 26, 2025May 29, 2025

Evaluation results

MethodLinks
DreamPRM (o4-mini)Model Access Type=Close-Source2025.0595.7
Kimi-k1.6-preview-20250308Model Access Type=Close-Source2025.0591.8
Step R1-V-MiniModel Access Type=Close-Source2025.0589.9
Doubao-pro-1.5Model Access Type=Close-Source2025.0588.9
AutoGPSModel Access Type=Open-Source, Base Model=InternVL32025.0586.2
Ovis2_34BModel Access Type=Open-Source2025.0584.6
Vision-R1-7BModel Access Type=Open-Source2025.0582.7
MathCoder-VL-8B#Params=8B2025.0573.6
InternVL2-76B#Params=76B2025.0567.8
Multimath-7B#Params=7B2025.0566.8
MathCoder-VL-2B#Params=2B2025.0566.4
GPT-4o#Params=-2025.0564.7
Claude3.5-Sonnet#Params=-2025.0564.4
MathGLM-Vision-9B#Params=9B2025.0564.4
MAVIS-7B#Params=7B2025.0564.1
IXC-2.5-Reward#Params=7B2025.0563.5
IXC-2-VL#Params=7B2025.0563
InternVL2-8B#Params=8B2025.0562
Gemini-1.5-Pro#Params=-2025.0558.9
GPT-4-turbo#Params=-2025.0558.3
Math-LLaVA-13B#Params=13B2025.0557.7
InternVL2-26B#Params=26B2025.0554.3
Claude3-Opus#Params=-2025.0552.9
GPT-4V#Params=-2025.0550.5
G-LLaVA-7B#Params=7B2025.0548.7
Human2025.0548.4
Human2025.0548.4
Math-PUMA-Qwen2#Params=8B2025.0548.1
Qwen-VL-Max#Params=-2025.0546.1
Qwen2-VL#Params=8B2025.0540.9
Math-PUMA-DS#Params=7B2025.0539.9
InternVL-Chat-2B-V1-5#Params=2B2025.0537.5
Qwen-VL-Plus#Params=-2025.0535.5
InfiMM-Math#Params=7B2025.0534.5
Deepseek-VL#Params=8B2025.0528.4
LLaVA-1.5-13B#Params=13B2025.0522.7
Random Chance2025.0521.6
SPHINX-V2-13B#Params=13B2025.0516.4