yeahhe365/AMC-WebUI

面向 Gemini 的 Local-First AI 工作流 WebUI，集成多模态聊天、Canvas、文件处理、实时搜索、代码执行与高级推理。

What it solves

AMC WebUI provides a comprehensive, local-first interface for interacting with AI models, primarily focusing on the native capabilities of Google Gemini while maintaining compatibility with OpenAI-standard chat APIs. It eliminates the need for multiple separate interfaces by consolidating deep reasoning, real-time audio/video, and advanced tool integration into a single console.

How it works

Built as a Vite + React SPA, the project follows a "Local-First" principle where chat data is stored in the browser's IndexedDB for privacy and speed. It can be deployed in three ways: as a standard static site, via Docker Compose (combining a frontend and a Node.js API proxy), or as a static frontend on Cloudflare Pages with a separate API backend. The system supports both direct browser-to-API communication (BYOK) and server-side key management for secure proxying.

Who it’s for

Users who want a powerful, privacy-centric UI for Gemini's advanced features (like Live API and Thinking models) and developers who need a unified console to manage multiple AI providers and MCP tools.

Highlights

  • Gemini Native Integration: Full support for Thinking (chain-of-thought visualization), Live API (real-time voice/vision), and Gemini Files API.
  • Local Python Sandbox: A browser-based Python environment using Pyodide (WASM) that automatically loads scientific libraries like numpy and pandas.
  • Model Context Protocol (MCP): Support for MCP via Node API, including human-in-the-loop authorization for tool calls.
  • Live Artifacts: Automatic rendering of code blocks into interactive HTML, SVG, Mermaid diagrams, and Graphviz charts.
  • Enterprise API Management: Dual-mode support for Gemini and OpenAI-compatible APIs with multi-key rotation and configuration isolation.
  • PWA Support: Installable as a desktop or mobile app with offline shell access and picture-in-picture mode.
  • Advanced File Handling: Browser-side audio preprocessing and support for ZIP/folder uploads to map codebase structures.

Related

  • Project
    Sophomoresty/gemini-web2api

    gemini‑web2api is a Python server that proxies Google Gemini’s private web API and presents it as an OpenAI‑compatible endpoint (`/v1/chat/completions`, `/v1/models`, etc.). It supports optional API‑key auth, streaming, function calling, image input, multiple Gemini model aliases, adjustable reasoning depth, and can run directly or inside Docker.

  • Project
    zhiyu1998/Gemi2Api-Server

    Gemi2Api‑Server is a FastAPI‑based bridge that lets you call Google Gemini via an OpenAI‑compatible REST API. It authenticates with Gemini using your browser cookies, forwards chat‑completion requests, optionally proxies generated images, and provides a simple admin UI. Deployable locally, via Docker, or on Render/HuggingFace, it’s useful for developers who want Gemini access without a formal public API.

  • Project
    Amm1rr/WebAI-to-API

    WebAI‑to‑API is a self‑hosted bridge that runs Google Gemini (via web API or a Playwright‑driven browser) and exposes it through OpenAI‑compatible endpoints, with a local dashboard and Docker support.

  • Project
    Nativu5/Gemini-FastAPI

    An OpenAI-compatible API wrapper for web-based Gemini models that allows free access using session cookies instead of an API key.

  • Project
    googleapis/js-genai

    Official Google npm SDK (`@google/genai`) that wraps Gemini’s public and Enterprise Agent APIs for JavaScript/TypeScript. Provides simple classes for one‑shot generation, streaming, stateful chats, file uploads, live sessions, and a unified Interactions API. Supports function/tool calling, automatic continuation, and multimodal inputs. Install with `npm i @google/genai`; use an API key or enterprise credentials. Documentation and samples are hosted on the repo’s website.