List your company
ResearchFormerly WizWand

Multimodal Reasoning on MM-Vista

81.4AccuracyGPT-5
49.99258.14666.374.454May 4, 2026May 13, 2026May 22, 2026Jun 1, 2026Jun 10, 2026Jun 19, 2026Jun 29, 2026
Updated 4mo ago

Evaluation results

MethodLinks
GPT-5Model Category=Closed-source models2026.0681.4
Gemini-2.5-ProModel Category=Closed-source models2026.0680.9
Qwen3-VL-8BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.0677.2
TACO (Ours-7B)Model Category=Code-tool / visual-agent models2026.0676.3
MathCoder-VL-8BModel Category=Code-tool / visual-agent models2026.0673.6
Qwen2.5-VL-32BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.0672.2
DeepEyes-v2-7BModel Category=Code-tool / visual-agent models2026.0671.9
CodeV-RL-7BModel Category=Code-tool / visual-agent models2026.0671.8
PyVision-RL-7BModel Category=Code-tool / visual-agent models2026.0671.4
Pixel-Reasoner-7BModel Category=Code-tool / visual-agent models2026.0671.2
InternVL3-8BModel Category=Open-source MLLMs (no visual tool)2026.0670.4
DeepEyes-7BModel Category=Code-tool / visual-agent models2026.0670.1
Thyme-7BModel Category=Code-tool / visual-agent models2026.0670
Qwen2.5-VL-7BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.0668.2
Mini-o3-7BModel Category=Code-tool / visual-agent models2026.0667.2
Unsilencing Latent ReasoningBackbone=VLAA Thinking-7B2026.0565.7
GPT-4oModel Category=Closed-source models2026.0663.4
DMLRBackbone=VLAA Thinking-7B2026.0562.9
ICoTBackbone=VLAA Thinking-7B2026.0561.4
VanillaBackbone=VLAA Thinking-7B2026.0561.1
CCoTBackbone=VLAA Thinking-7B2026.0560.5
MCoTBackbone=VLAA Thinking-7B2026.0559.6
LLaVA-OneVision-7BModel Category=Open-source MLLMs (no visual tool)2026.0658.6
DMLRBackbone=R1 OneVision-7B2026.0558
Unsilencing Latent ReasoningBackbone=R1 OneVision-7B2026.0557.3
ICoTBackbone=R1 OneVision-7B2026.0555.6
CCoTBackbone=R1 OneVision-7B2026.0553.4
MCoTBackbone=R1 OneVision-7B2026.0552.5
VanillaBackbone=R1 OneVision-7B2026.0551.2