List your company
ResearchFormerly WizWand

Multimodal Reasoning on LVista

73.8AccuracyGemini-2.5-Pro
31.6842.61553.5564.485Jun 29, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Gemini-2.5-ProModel Category=Closed-source models2026.0673.8
GPT-5Model Category=Closed-source models2026.0669.3
TACO (Ours-7B)Model Category=Code-tool / visual-agent models2026.0655.6
Qwen3-VL-8BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.0654.9
Qwen2.5-VL-32BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.0654.4
GPT-4oModel Category=Closed-source models2026.0653.2
PyVision-RL-7BModel Category=Code-tool / visual-agent models2026.0649.2
Thyme-7BModel Category=Code-tool / visual-agent models2026.0649
DeepEyes-v2-7BModel Category=Code-tool / visual-agent models2026.0648.7
CodeV-RL-7BModel Category=Code-tool / visual-agent models2026.0648.7
DeepEyes-7BModel Category=Code-tool / visual-agent models2026.0647.7
Pixel-Reasoner-7BModel Category=Code-tool / visual-agent models2026.0646.4
Mini-o3-7BModel Category=Code-tool / visual-agent models2026.0645.7
InternVL3-8BModel Category=Open-source MLLMs (no visual tool)2026.0645.6
MathCoder-VL-8BModel Category=Code-tool / visual-agent models2026.0641.6
Qwen2.5-VL-7BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.0639.8
LLaVA-OneVision-7BModel Category=Open-source MLLMs (no visual tool)2026.0633.3