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

Multimodal Reasoning on MathVerse

63.8Accuracy

OpenMMReasoner-7B

Updated 7d ago

Evaluation Results

Method
OpenMMReasoner-7B 2026.05	63.8	-
AnE-3rd 2026.05	62.3	-
AnE-2nd 2026.05	60.9	-
ADHint 2026.05	60.6	-
AnE-1st 2026.05	60.5	-
NPO 2026.04	60	-
GRPO 2026.04	59.52	-
NPO 2026.04	59.38	-
AutoNPO 2026.04	59.11	-
RLEP 2026.04	58.91	-
ExGRPO 2026.04	56.89	-
Qwen2.5-VL-72B w/ RAPID 2025.06	56.2	-
PyVision-Image 2026.02	55.8	-
SRPO 2026.05	55.8	-
LUFFY 2026.04	55.58	-
MMR1 2026.05	55.4	-
TGRL-DAPO 2026.03	55.21	-
Qwen3-VL-8B-Instruct 2026.04	54.61	-
VL-Rethinker-72B 2025.06	54.6	-
Athena-PRM 2025.06	54.6	-
Qwen2.5-VL-32B w/ RAPID 2025.06	54.4	-
Qwen2.5-VL-32B w/ RAPID 2025.06	54.3	-
Perception-R1 2026.03	54.3	-
VL-Rethinker 2026.05	54.2	-
Revisual-R1 2026.05	53.6	-
Preliminary RL 2026.05	53.6	-
TGRL-DAPO 2026.03	53.51	-
NoisyRollout 2026.03	53	-
SPARK-VL-7B 2026.05	53	-
DeepEyes-v2 2026.02	52.7	-
DeepEyesV2 2025.11	52.7	-
ReVisual-R1-7B 2025.06	52.7	-
TGRL-GRPO 2026.03	52.5	-
Vision-R1 2026.05	52.4	-
Qwen2.5-VL-7B w/ RAPID 2025.06	52.2	-
MM-Eureka-32B 2025.06	51.5	-
Athena-ORM 2025.06	51.2	-
TGRL-GRPO 2026.03	51.08	-
InternVL3-78B 2025.06	51	-
Metis-RISE 2026.05	51	-
GRPO 2026.03	50.8	-
DAPO 2026.03	50.8	-
Self-consistency 2025.06	50.8	-
DAPO 2026.03	50.6	-
GRPO 2026.03	50.46	-
MM-Eureka-Qwen-7B 2025.12	50.3	-
ThinkLite 2026.03	50.2	-
Ovis2-34B 2025.06	50.1	-
Qwen2.5-VL-32B 2025.06	50.1	-
VLAA-Thinker 2026.03	49.9	-
MSSR 2025.12	49.8	-
LLaVA-Critic-R1 2026.05	49.7	-
Qwen2.5-VL-7B-Instruct 2026.05	49.2	-
GPT-4.1 2025.09	48.9	-
Bagel-Zebra-CoT-7B 2025.09	48.8	-
Qwen2.5-VL-3B w/ RAPID 2025.06	48.6	-
GRPO 2025.12	48.5	-
VLAA-Thinker-7B 2025.12	48.2	-
InternVL3-38B 2025.06	48.2	-
QVQ-72B-Preview 2025.06	48.2	-
OmniCaptioner 2025.06	48	-
OpenVLThinker 2026.03	48	-
OpenVLThinker 2026.05	47.9	-
RLOO 2025.12	47.8	-
Gemini-2.0-Flash 2025.06	47.7	-
VL-Rethinker-7B 2025.06	47.5	-
ThinkLite-VL-7B 2025.12	47.3	-
DeepEyes 2026.02	47.3	-
DeepEyes 2025.11	47.3	-
Qwen2.5-VL-72B 2025.06	47.3	-
DeepEyes-7B 2025.06	47.3	-
Qwen2.5-VL-72B 2025.06	47.3	-
Qwen2.5-VL-72B 2025.06	47.3	-
CodeDance 2026.02	46.8	-
Qwen2.5-VL-7B (Bo8) 2025.06	46.8	-
Claude-3.7-Sonnet 2025.06	46.7	-
Claude3.7-Sonnet 2025.09	46.7	-
CoT Cold-Start + Solver Feedback 2025.11	46.65	-
Self-Instruct 2025.11	46.4	-
Athena-PRM 2025.06	46.3	-
Qwen-2.5-VL-7B 2026.03	46.2	-
Self-Instruct + Solver Feedback 2025.11	46.17	-
MM-Eureka-7B 2025.06	46.1	-
R1-Onevision 2026.03	46.1	-
R1-Onevision-7B 2025.12	46	-
OpenVLThinker-7B 2025.12	45.8	-
Qwen2.5-VL-7B 2025.12	45.8	-
Ovis2-16B 2025.06	45.8	-
Qwen2.5-VL-7B 2026.02	45.6	-
Qwen-2.5-VL 2025.11	45.6	-
Self-Instruct + CoT-Self-Instruct 2025.11	45.58	-
CoT Cold-Start 2025.11	45.41	-
Seed Set 2025.11	45.23	-
Self-Instruct + R-Zero 2025.11	44.72	-
InternVL3-14B 2025.06	44.4	-
VisualPRM-8B 2025.06	44.3	-
Athena-ORM 2025.06	44.1	-
Self-consistency 2025.06	43.7	-
Base Model 2025.11	43.53	-
DeepSketcher-7B 2025.09	43.2	-

Showing 100 of 141 rows