Open source implementation and extension of Google Research’s PaperBanana for automated academic figures, diagrams, and research visuals, expanded to new domains like slide generation.

PaperBanana-CN：支持各类中转站的科研生图工具（中文友好，包含最新gpt-image-2）

AI agent for civil engineering research papers — 15 journals, 18 utilities, 4 templates, CLI + Web UI + Docker, PaperBanana diagrams, 245 tests. v1.0.0 released.

PaperBanana-AIMSLab是一个面向科研写作场景的AI学术配图助手，基于开源项目PaperBanana优化而来。它可以根据论文方法章节、图注或图表意图自动生成高质量学术配图，并支持多 Agent 规划、风格优化、图像生成、评审迭代和图片精修。项目针对国内使用环境增强了多 API 通道自动切换能力，支持 AIPAIBOX、Google Gemini、OpenAI 与兼容网关，并新增gpt-image-2图像生成与image-to-image精修支持。

🍌 OpenClaw skill for generating publication-quality academic diagrams and plots from text using PaperBanana

PaperBananaFX is a JavaFX desktop orchestration demo for a real-time AI workflow. It targets a specific bottleneck in the "AI scientist" workflow: converting method text and figure captions into publication-ready illustrations with semantic fidelity to the method and stylistic conformity to modern research-paper aesthetics.