{"slug": "oss-chatgpt-webui-v4-projects-agent-profiles-server-tools-publishing", "title": "OSS ChatGPT WebUI v4 – Projects, Agent Profiles, Server Tools, Publishing", "summary": "The self-hosted, OpenAI-compatible AI gateway llms.py released version 4 on July 20, 2026, adding projects, agent profiles, server tools, and publishing capabilities. The update supports 530+ models from 24 providers and introduces an extensions system, Gemini file search, and credential authentication.", "body_md": "# llms.py\n\nThe self-hosted, OpenAI-compatible AI gateway for text, image & audio — a lightweight CLI, server and OSS Open WebUI alternative for Local and Cloud LLMs\n\nJuly 20, 2026 - [v4 Released! →](/docs/latest)\n\nOne tool, every modality\n\n## Generate Anything\n\nFrom a single self-hosted gateway — chat with 530+ models, create images, and synthesize speech across 24 providers\n\n[Text & Chat](/docs/features/chat-ui)\n\nA fast, private, ChatGPT-like web UI over every local & cloud LLM — with tools, MCP, skills, agents and 200+ system prompts\n\n[Image Generation](/docs/media-generation/image-generation)\n\nCreate images via Gemini, OpenAI, OpenRouter, Chutes, Z.ai & Nvidia — right inside the chat and gallery workflow\n\n[Audio & Speech](/docs/media-generation/audio-generation)\n\nHigh-quality text-to-speech with Gemini & OpenRouter models, plus voice-to-text input via the microphone\n\nCreate → Publish → Showcase\n\n## Your Creations, in the Public Gallery\n\nAnything you generate in llms.py can be published with one click — landing in the live showcase at [ai.llmspy.org](https://ai.llmspy.org), browsable by everyone with categories, tags, ratings and infinite scroll.\n\n- Browse generated images & audio by category and tag\n- Play full\n[Projects built by AI](#projects-built)— games and apps, ready to run - Free & anonymous publisher account — no email required\n\nPlayable & self-contained\n\n## Projects Built by AI\n\nComplete, self-contained apps & games — generated with llms.py and published for anyone to open and play\n\n### Quick Install\n\n`pip install llms-py`\n\n### 💬 Simple Chat\n\nAsk questions directly from the command line\n\n### Run Server\n\n`llms --serve 8000`\n\n## Beautiful Themes\n\nPersonalize your experience with a selection of[handcrafted themes](/docs/features/themes)\n\n### 💬 ChatGPT-like Interface\n\nModern, fast, and privacy-focused web UI for all your local and remote LLMs\n\n## What's New\n\nLatest release focused on extensibility, expanded provider support, and enhanced user experience\n\n[530+ Models from 24 Providers](/docs/v3#switch-to-modelsdev-provider-model-configuration)\n\nPowered by models.dev integration with automatic daily updates\n\n[Extensions System](/docs/extensions)\n\nAdd features, providers, and customize the UI with flexible plugins\n\n[Gemini File Search](/docs/extensions/gemini)\n\nRAG workflows with document stores, categories, and contextual chat\n\n[Credentials Auth](/docs/authentication/credentials)\n\nUsername/Password authentication with Admin UI and CLI user management\n\n[Agent Profiles](/docs/features/agents)\n\nSpecialized AI agents with custom prompts, tools, themes, and Planner→Coder workflows\n\n[Projects](/docs/features/projects)\n\nSecure workspace isolation that restricts AI agent filesystem access to designated project folders\n\n[Tool Support](/docs/extensions/tools)\n\nFirst-class Python function calling for LLM interactions with your environment\n\n[Server Tools](/docs/features/server-tools)\n\nBuilt-in support for provider hosted Anthropic & OpenRouter server tools like web search\n\n[MCP Support](/docs/mcp/fast_mcp)\n\nConnect to Model Context Protocol servers for extended tool capabilities\n\n[Skills Support](/docs/extensions/skills)\n\nExtend AI capabilities with specialized knowledge, workflows, and tools\n\n[Agent Browser](/docs/features/browser)\n\nAI-powered browser automation with live preview, script editor, and element inspector\n\n[Themes](/docs/features/themes)\n\nCustomize the look and feel with built-in themes or create your own\n\n[Calculator & Run Code UIs](/docs/features/calculator-ui)\n\nBeautiful UIs to evaluate math and execute Python, JS, TS & C# code\n\n[Media Gallery](/docs/media-generation/media-gallery)\n\nBrowse and manage all your generated images and audio in one place\n\n[Voice Input](/docs/features/voice-input)\n\nVoice-to-text transcription via microphone button or ALT+D keyboard shortcut\n\n[Image Generation](/docs/media-generation/image-generation)\n\nBuilt-in support for Gemini, OpenAI, OpenRouter, Chutes, Z.ai and Nvidia\n\n[Audio Generation](/docs/media-generation/audio-generation)\n\nTTS support for Gemini 2.5 Flash/Pro Preview models\n\n[KaTeX Math Rendering](/docs/features/katex)\n\nBeautiful LaTeX math typesetting for equations and formulas\n\n## Model Selector\n\nSmart search, advanced filtering, sorting, and favorites over 530 models from 24 providers\n\n[Learn more →](/docs/features/model-selector)\n\n## Credentials Auth\n\nBuilt-in Username/Password authentication with a Sign In page, Admin Web UI and CLI for user managing accounts, roles, and account locking\n\n[Learn more →](/docs/authentication/credentials)\n\n## Agent Profiles & Projects\n\nConfigure specialized AI agents with custom prompts, tools, and themes — and scope their filesystem access to secure project workspaces\n\n## Gemini File Search & RAG\n\nBuild knowledge bases with document stores and contextual AI chat\n\n[Learn more →](/docs/extensions/gemini)\n\n### Server Tools\n\nBuilt-in support for provider-hosted OpenRouter & Anthropic server tools like web search, web fetch & code execution\n\n[Learn more →](/docs/features/server-tools)\n\n## Agent Browser\n\nAn integrated workspace for building and running automated browser scripts with AI assistance, live previews, and an interactive element inspector\n\n[Learn more →](/docs/features/browser)\n\n## Image & Audio Generation\n\nSeamless media generation through UI and CLI\n\n## Runtime Provider Management\n\nEnable or disable providers on the fly without configuration changes\n\n[Learn more →](/docs/configuration)\n\n### Ready to Get Started?\n\nInstall llms.py and start chatting with 530+ AI models in minutes", "url": "https://wpnews.pro/news/oss-chatgpt-webui-v4-projects-agent-profiles-server-tools-publishing", "canonical_source": "https://llmspy.org", "published_at": "2026-07-20 16:03:46+00:00", "updated_at": "2026-07-20 16:26:22.779515+00:00", "lang": "en", "topics": ["artificial-intelligence", "generative-ai", "ai-tools", "ai-products", "developer-tools"], "entities": ["llms.py", "OpenAI", "Gemini", "OpenRouter", "Chutes", "Z.ai", "Nvidia", "Anthropic"], "alternates": {"html": "https://wpnews.pro/news/oss-chatgpt-webui-v4-projects-agent-profiles-server-tools-publishing", "markdown": "https://wpnews.pro/news/oss-chatgpt-webui-v4-projects-agent-profiles-server-tools-publishing.md", "text": "https://wpnews.pro/news/oss-chatgpt-webui-v4-projects-agent-profiles-server-tools-publishing.txt", "jsonld": "https://wpnews.pro/news/oss-chatgpt-webui-v4-projects-agent-profiles-server-tools-publishing.jsonld"}}