{
  "markdown": "<!-- mcp-name: io.github.mituan-ai/paperbanana-cn -->\n\n<p align=\"right\">\n  <strong>English</strong> ·\n  <a href=\"https://github.com/mituan-ai/PaperBanana-CN/blob/main/README_CN.md\">简体中文</a>\n</p>\n\n<p align=\"center\">\n  <img\n    src=\"https://raw.githubusercontent.com/mituan-ai/PaperBanana-CN/main/assets/readme/hero.webp\"\n    width=\"100%\"\n    alt=\"PaperBanana-CN scientific figure workbench with a generated multimodal fault-diagnosis figure\"\n  >\n</p>\n\n<p align=\"center\">\n  PaperBanana-CN generates methodology diagrams from research descriptions and statistical plots\n  from CSV or JSON data. It uses the PaperBanana scientific workflow and adds separate VLM and\n  image-service connections, a Chinese Studio interface, and explicit aspect-ratio and resolution\n  controls.\n</p>\n\n<p align=\"center\">\n  <a href=\"#quick-start\"><img src=\"https://img.shields.io/badge/LAUNCH_STUDIO-uvx_paperbanana--cn_studio-147862?style=for-the-badge&logo=gnometerminal&logoColor=white\" alt=\"Launch Studio\"></a>\n  <a href=\"https://pypi.org/project/paperbanana-cn/\"><img src=\"https://img.shields.io/badge/INSTALL-PYPI-3775A9?style=for-the-badge&logo=pypi&logoColor=white\" alt=\"Install from PyPI\"></a>\n</p>\n\n<p align=\"center\">\n  <a href=\"https://github.com/mituan-ai/PaperBanana-CN/actions/workflows/ci.yml\"><img src=\"https://img.shields.io/github/actions/workflow/status/mituan-ai/PaperBanana-CN/ci.yml?branch=main&style=flat-square&logo=githubactions&logoColor=white&label=CI\" alt=\"CI status\"></a>\n  <img src=\"https://img.shields.io/badge/Package-2.0.1-147862?style=flat-square\" alt=\"Package version 2.0.1\">\n  <a href=\"https://github.com/mituan-ai/PaperBanana-CN/blob/main/LICENSE\"><img src=\"https://img.shields.io/badge/License-MIT-52605B?style=flat-square\" alt=\"MIT License\"></a>\n</p>\n\n## One run from input to figure\n\n<p align=\"center\">\n  <img\n    src=\"https://raw.githubusercontent.com/mituan-ai/PaperBanana-CN/main/assets/readme/studio-workflow.gif\"\n    width=\"960\"\n    alt=\"A PaperBanana-CN Studio run progressing from configured inputs to a completed scientific figure\"\n  >\n</p>\n\nThe recording follows a methodology-diagram run from submitted inputs to the completed result.\n\n## Before you start\n\nYou need Python 3.10-3.12, [uv](https://docs.astral.sh/uv/), and a desktop browser.\n\n| Task | Required model connections |\n|---|---|\n| Methodology diagram | VLM and image generation |\n| Statistical plot | VLM only |\n| Multi-panel composition and run browsing | None |\n\nEach connection specifies its own protocol, Base URL, API key, model name, and timeout. The VLM and\nimage roles may use the same service or two different services.\n\n## Quick start\n\n### 1. Launch Studio\n\n```bash\nuvx paperbanana-cn studio\n```\n\nOpen <http://127.0.0.1:7860>. `uvx` runs the package in an isolated environment and does not modify\nDebian or Ubuntu's system Python.\n\n### 2. Add the model connections\n\nOpen **Settings → VLM connection**, fill in the service fields, and select **Save and use**. Repeat\nunder **Image connection** before generating a methodology diagram.\n\nEditing a saved connection does not activate it. An empty API-key field keeps the stored key.\nStudio does not fill stored keys back into the browser.\n\n<details>\n<summary><strong>Connection manager screenshot and protocol notes</strong></summary>\n\n<br>\n\n<p align=\"center\">\n  <img\n    src=\"https://raw.githubusercontent.com/mituan-ai/PaperBanana-CN/main/assets/readme/connections-zh.webp\"\n    width=\"100%\"\n    alt=\"PaperBanana-CN connection manager with separate fields for a VLM service\"\n  >\n</p>\n\nThe [connection guide](https://github.com/mituan-ai/PaperBanana-CN/blob/main/docs/CONNECTIONS.md)\nlists the supported protocols, credential storage rules, connection tests, and legacy mode.\n\n</details>\n\n### 3. Generate a figure\n\nOpen **Methodology diagram**, provide the method content and figure caption, then choose an aspect\nratio, resolution, and output format.\n\nThe same task from the CLI:\n\n```bash\npaperbanana-cn generate \\\n  --input method.txt \\\n  --caption \"Overview of the proposed architecture\" \\\n  --aspect-ratio 16:9 \\\n  --resolution 2K \\\n  --format png\n```\n\n## What PaperBanana-CN adds\n\n### Separate VLM and image connections\n\nThe two model roles have independent protocol, Base URL, API key, model, and timeout settings.\nStudio, CLI, and MCP resolve the same saved connections. Saved profiles contain credential\nreferences; API keys remain outside the repository and run metadata.\n\nOfficial APIs, OpenAI-compatible services, and Gemini-compatible services are supported. Provider\nspecifics stay in the adapters rather than the scientific workflow.\n\n### Chinese and English Studio\n\nThe Studio interface, help text, validation, progress messages, and errors are available in Chinese\nand English. Changing the interface language does not rewrite prompts, paper text, or labels inside\nthe generated figure.\n\n### Aspect ratios and resolution\n\nSupported aspect ratios:\n\n`1:1` · `4:3` · `3:2` · `5:4` · `16:9` · `21:9` · `4:5` · `3:4` · `2:3` · `9:16`\n\nResolution tiers:\n\n`1K` · `2K` · `4K`\n\nStudio shows the request size or provider-native tier before generation. If an adapter cannot\nproduce the selected combination, validation stops the request and reports the unsupported option.\n\n## Studio workflows\n\n| Area | Workflow | Model connections |\n|---|---|---|\n| Create | Methodology diagram | VLM and image |\n| Create | Statistical plot | VLM |\n| Improve | Continue a saved run | Depends on the saved run |\n| Improve | Quality evaluation | VLM |\n| Automate | Full-paper orchestration | VLM and image |\n| Automate | Batch generation | Depends on the task type |\n| Automate | Parameter sweep | VLM and image |\n| Tools | Multi-panel composition | None |\n| Tools | Run browser | None |\n\n<p align=\"center\">\n  <img\n    src=\"https://raw.githubusercontent.com/mituan-ai/PaperBanana-CN/main/assets/readme/studio-methodology-en.webp\"\n    width=\"100%\"\n    alt=\"PaperBanana-CN methodology-diagram workspace with task inputs and a completed result\"\n  >\n</p>\n\n<p align=\"center\"><sub>Methodology-diagram workspace</sub></p>\n\n<p align=\"center\">\n  <img\n    src=\"https://raw.githubusercontent.com/mituan-ai/PaperBanana-CN/main/assets/readme/studio-statistical-plot-en.webp\"\n    width=\"100%\"\n    alt=\"PaperBanana-CN statistical-plot workspace with CSV input and a completed line chart\"\n  >\n</p>\n\n<p align=\"center\"><sub>Statistical-plot workspace using synthetic demonstration data</sub></p>\n\n## CLI, MCP, Docker, and Colab\n\n| Entry point | Command or link |\n|---|---|\n| Studio | `paperbanana-cn studio` |\n| CLI | `paperbanana-cn generate --help` |\n| MCP server | `paperbanana-cn mcp` |\n| GitHub Action | [Action reference](https://github.com/mituan-ai/PaperBanana-CN/blob/main/integrations/github-action/README.md) |\n| Docker | `ghcr.io/mituan-ai/paperbanana-cn:2.0.1` |\n| Colab | [Quickstart notebook](https://colab.research.google.com/github/mituan-ai/PaperBanana-CN/blob/main/notebooks/PaperBanana_CN_Quickstart.ipynb) |\n\n<details>\n<summary><strong>MCP client configuration</strong></summary>\n\n```json\n{\n  \"mcpServers\": {\n    \"paperbanana-cn\": {\n      \"command\": \"uvx\",\n      \"args\": [\"paperbanana-cn\", \"mcp\"]\n    }\n  }\n}\n```\n\nThe server provides 11 tools and reads the same active connections as Studio and CLI. See the\n[MCP guide](https://github.com/mituan-ai/PaperBanana-CN/blob/main/mcp_server/README.md) for the tool\nlist and arguments.\n\n</details>\n\n<details>\n<summary><strong>Docker</strong></summary>\n\n```bash\ndocker run --rm -p 7860:7860 \\\n  -v paperbanana-cn-config:/home/paperbanana/.config/paperbanana-cn \\\n  -v paperbanana-cn-data:/home/paperbanana/.local/share/paperbanana-cn \\\n  -v paperbanana-cn-outputs:/work/outputs \\\n  ghcr.io/mituan-ai/paperbanana-cn:2.0.1 \\\n  studio --host 0.0.0.0\n```\n\n</details>\n\n<details>\n<summary><strong>Permanent install, source setup, and optional providers</strong></summary>\n\nInstall the command in a uv-managed environment:\n\n```bash\nuv tool install paperbanana-cn\npaperbanana-cn studio\n```\n\nRun the current source checkout:\n\n```bash\ngit clone https://github.com/mituan-ai/PaperBanana-CN.git\ncd PaperBanana-CN\nuv sync\nuv run paperbanana-cn studio\n```\n\nThe default package includes Studio, MCP, PDF input, OpenAI-compatible services, and Gemini.\n\n| Optional adapter | Install |\n|---|---|\n| AWS Bedrock | `uv tool install \"paperbanana-cn[bedrock]\"` |\n| Anthropic | `uv tool install \"paperbanana-cn[anthropic]\"` |\n| LiteLLM | `uv tool install \"paperbanana-cn[litellm]\"` |\n| All optional providers | `uv tool install \"paperbanana-cn[all-providers]\"` |\n\nFor CI, use `paperbanana-cn connections add --api-key-env ENV_VAR` so the key is read from an\nenvironment variable instead of a command-line value.\n\n</details>\n\n## V1 and upstream\n\nV2 is maintained on `main` as the `paperbanana-cn` distribution, the `paperbanana_cn` Python\nmodule, and the `paperbanana-cn` command.\n\n- [Browse the frozen `v1` branch](https://github.com/mituan-ai/PaperBanana-CN/tree/v1)\n- [Download the `v1.0.0` release](https://github.com/mituan-ai/PaperBanana-CN/releases/tag/v1.0.0)\n\nThe scientific figure-generation core is based on\n[`llmsresearch/paperbanana`](https://github.com/llmsresearch/paperbanana). PaperBanana-CN is an\nunofficial community implementation and is not affiliated with or endorsed by the upstream\nauthors.\n\n## Community\n\nPaperBanana-CN is maintained by [mituan](https://github.com/mituan-ai) under the\n[MIT License](https://github.com/mituan-ai/PaperBanana-CN/blob/main/LICENSE).\n\n- Ask usage questions in [Discussions](https://github.com/mituan-ai/PaperBanana-CN/discussions).\n- Report reproducible bugs in [Issues](https://github.com/mituan-ai/PaperBanana-CN/issues).\n- Report vulnerabilities through [Private Vulnerability Reporting](https://github.com/mituan-ai/PaperBanana-CN/security/advisories/new).\n- Read [CONTRIBUTING.md](https://github.com/mituan-ai/PaperBanana-CN/blob/main/CONTRIBUTING.md) before opening a pull request.\n\n<details>\n<summary><strong>Development checks</strong></summary>\n\n```bash\ngit clone https://github.com/mituan-ai/PaperBanana-CN.git\ncd PaperBanana-CN\nuv sync --extra dev\nuv run pytest tests/ -q\nuv run ruff check paperbanana_cn/ mcp_server/ tests/ scripts/\n```\n\nDo not upload API keys, private relay URLs, unpublished papers, private datasets, local connection\nstores, or generated run directories.\n\n</details>\n\n## Star history\n\n<a href=\"https://www.star-history.com/?repos=mituan-ai%2FPaperBanana-CN&type=date&legend=top-left\">\n <picture>\n   <source media=\"(prefers-color-scheme: dark)\" srcset=\"https://api.star-history.com/chart?repos=mituan-ai/PaperBanana-CN&type=date&theme=dark&legend=top-left&sealed_token=tONDl7QT6gBodlxbICyg-BGsu060cE2rb7tZmOubJS6r7ZQMt8tGi9pUE274ujDrVgxHmy3U6QwUFtqtCDbU5abOpd8t9gKCK6B48Typy5z9FLLBvnF4uA\" />\n   <source media=\"(prefers-color-scheme: light)\" srcset=\"https://api.star-history.com/chart?repos=mituan-ai/PaperBanana-CN&type=date&legend=top-left&sealed_token=tONDl7QT6gBodlxbICyg-BGsu060cE2rb7tZmOubJS6r7ZQMt8tGi9pUE274ujDrVgxHmy3U6QwUFtqtCDbU5abOpd8t9gKCK6B48Typy5z9FLLBvnF4uA\" />\n   <img alt=\"Star History Chart\" src=\"https://api.star-history.com/chart?repos=mituan-ai/PaperBanana-CN&type=date&legend=top-left&sealed_token=tONDl7QT6gBodlxbICyg-BGsu060cE2rb7tZmOubJS6r7ZQMt8tGi9pUE274ujDrVgxHmy3U6QwUFtqtCDbU5abOpd8t9gKCK6B48Typy5z9FLLBvnF4uA\" />\n </picture>\n</a>\n",
  "bytes": 11296,
  "sha": "bb5ea4d2600cf410f39588319164b0c545365ef89325dc17d7fdbb5a10974678",
  "repo_slug": "mituan-ai/paperbanana-cn",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_mituan_ai_paperbanana_cn_d6829f4d/readme"
}