RealWorldQA
RealWorldQA environment for evaluating vision-language models on real-world question answering
- Domain
- rl-env
- License
- unknown
- Published
- Oct 2025
Cite
Notes
Only stored in your browser.
Top score 60.0% by Gemini 2.5 Flash - 4 models reporting
Score history
2Top models
4
Related tools
1Implementations, trainers, datasets and scaffolds linked to this eval.
FAQ
- What is RealWorldQA?
- RealWorldQA environment for evaluating vision-language models on real-world question answering
- What is the current top score on RealWorldQA?
- The top reported score is 60.0% by Gemini 2.5 Flash, across 4 models reporting.
- How can a model improve its RealWorldQA score?
- Tools linked to RealWorldQA on Sophon include Realworldqa RL Env (Community) - RL environments, datasets, and scaffolds that target this eval.
- What license is RealWorldQA under?
- RealWorldQA is available under unknown.