Miffan

More about this project

OpenAI Chat Completions and Responses-compatible services, Google Gemini and Vertex AI, and Anthropic Claude-compatible services - Browser-based OpenAI Codex sign-in using access included with eligible ChatGPT subscriptions - Built-in presets for popular official services and gateways, plus fully custom providers, base URLs, models, request paths, headers, and body parameters - Model discovery and configurable modality, reasoning, tool-use, context-window, and generation settings - Authenticated HTTP/SOCKS5 proxy support, custom User-Agent, connection testing, and optional provider balance queries - Chat, reasoning, tool calls, image generation, and multimodal input according to the selected model's capabilities

Streaming responses, message editing and regeneration, response branches, favorites, folders, and local full-text search - Per-conversation system prompts, history compression, automatic titles, follow-up suggestions, token usage, and generation statistics - Images and document attachments; local text extraction for PDF, DOCX, PPTX, and EPUB when needed - Rich Markdown and HTML rendering with syntax highlighting, LaTeX, tables, Mermaid diagrams, images, and diffs - Export conversations as Markdown or images, share content into Miffan, and hand shared text to a selected assistant

Independent assistant configuration for models, prompts, sampling parameters, context…