OpenCodePapers

visual-question-answering-vqa-on-vibe-eval

Visual Question Answering (VQA)
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeAccuracyModelNameReleaseDate
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities67.2Gemini 2.5 Pro2025-07-07
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities65.4Gemini 2.5 Flash2025-07-07
Introducing Gemini 2.0: our new AI model for the agentic era56.3Gemini 2.0 Flash Experimental2024-12-11
LLaVA-OneVision: Easy Visual Task Transfer✓ Link50.7LLaVA-OneVision-72B2024-08-06