List your company
ResearchFormerly WizWand

Multi-modal Reasoning on MathVision (test)

47.7Accuracy (%)Nanobanana-Pro
15.04423.5223240.478Dec 8, 2025Dec 21, 2025Jan 3, 2026Jan 17, 2026Jan 30, 2026Feb 12, 2026Feb 26, 2026
Updated 7mo ago

Evaluation results

MethodLinks
Nanobanana-ProData Filtration=Filt2026.0147.7
Qwen-ImageData Filtration=Filt2026.0147
Qwen-ImgcoderSynthesis Strategy=ImgCoder, Base Model=Qwen2026.0146.9
Gemini-ImgCoderSynthesis Strategy=ImgCoder, Base Model=Gemini2026.0146.9
Nanobanana-Pro2026.0146.1
Qwen-Image2026.0145.9
Gemini-2.0-Flash2026.0241.3
Base2026.0139
Qwen2.5-VL-72BModel Category=General Multimodal LLM, Parameters=72B2025.1238.1
SFT-MModel=Qwen2.5-VL-7B, Paradigm=SFT-M2026.0232.96
SRPO-7BModel Category=Reasoning Multimodal LLM, Parameters=7B2025.1232.9
ThinkOmniBase Model=Qwen2.5-Omni-7B, LRM Guide=Qwen32026.0232.9
GRPOModel=Qwen2.5-VL-7B, Paradigm=GRPO2026.0232.83
VL-Rethinker-7BModel Category=Reasoning Multimodal LLM, Parameters=7B2025.1232.3
SFT-RSModel=Qwen2.5-VL-7B, Paradigm=SFT-RS2026.0232.25
SFTModel=Qwen2.5-VL-7B, Paradigm=SFT2026.0232.11
SFT-MModel=Qwen2.5-VL-3B, Paradigm=SFT-M2026.0231.92
ThinkOmniBase Model=Omni-R1-7B, LRM Guide=Qwen32026.0231.5
GPT-4o2026.0230.4
ReLaX-VL-7BModel Category=Reasoning Multimodal LLM, Parameters=7B2025.1230.2
GRPOModel=Qwen2.5-VL-3B, Paradigm=GRPO2026.0229.96
SFT-RSModel=Qwen2.5-VL-3B, Paradigm=SFT-RS2026.0229.83
SFTModel=Qwen2.5-VL-3B, Paradigm=SFT2026.0229.5
ThinkOmniBase Model=Qwen2.5-Omni-7B, LRM Guide=DeepSeek2026.0228.2
ReLaX-VL-3BModel Category=Reasoning Multimodal LLM, Parameters=3B2025.1227.6
Vision-R1-7BModel Category=Reasoning Multimodal LLM, Parameters=7B2025.1227.2
ThinkOmniBase Model=Omni-R1-7B, LRM Guide=DeepSeek2026.0227
MM-Eureka-7BModel Category=Reasoning Multimodal LLM, Parameters=7B2025.1226.9
Omni-R1Notes=Evaluated using authors' scripts2026.0225.4
HumanOmniV2Notes=Evaluated using authors' scripts2026.0225.4
Omni-R12026.0225.4
OpenVLThinker-7BModel Category=Reasoning Multimodal LLM, Parameters=7B2025.1225.3
ThinkOmniBase Model=Qwen2.5-Omni-3B, LRM Guide=Qwen32026.0225.3
Qwen2.5-VL-7BModel Category=General Multimodal LLM, Parameters=7B2025.1225.1
Qwen2.5-Omni-7B2026.0225
R1-VL-7BModel Category=Reasoning Multimodal LLM, Parameters=7B2025.1224.7
MM-Eureka-8BModel Category=Reasoning Multimodal LLM, Parameters=8B2025.1222.2
Kimi-VL-16BModel Category=General Multimodal LLM, Parameters=16B2025.1221.4
Qwen2.5-VL-3BModel Category=General Multimodal LLM, Parameters=3B2025.1221.2
ThinkOmniBase Model=Qwen2.5-Omni-3B, LRM Guide=DeepSeek2026.0220.2
Intern2.5-VL-8BModel Category=General Multimodal LLM, Parameters=8B2025.1219.7
Llava-OV-7BModel Category=General Multimodal LLM, Parameters=7B2025.1218.5
Qwen2.5-Omni-3B2026.0218.2
Intern2-VL-8BModel Category=General Multimodal LLM, Parameters=8B2025.1217.4
Qwen2-VL-7BModel Category=General Multimodal LLM, Parameters=7B2025.1216.3