BREAKING
PaperBanana Turns Text Into Diagrams
Built on Google Research Paper
0
GitHub stars
0
specialized agents
0$
per image (Gemini)
Two-Phase Agent Pipeline
0Input Optimization
1Linear Planning
2Iterative Refinement
Swappable Default Models
DefaultOpenAI
VLM: gpt-5.2
Image: gpt-image-1.5
Low-costGoogle
VLM: gemini-2.5-flash
Image: gemini-3-pro-image
More Verification Expected
AI NEWS BLITZ
An open-source tool now generates academic figures straight from text descriptions.