Text Generation Web UI

Text Generation Web UI — the oobabooga textgen app — is the most popular open-source desktop application for running local LLMs, with 47,700+ stars on GitHub.

Overview

Often described as the AUTOMATIC1111 of text generation, Text Generation Web UI gives local models a polished home. It loads GGUF and other model formats through llama.cpp, Transformers and ExLlamaV2 backends, and offers chat, notebook and API modes in one interface: roleplay-style character chats, instruct conversations, image input for vision models, tool calling, and a large extensions ecosystem for everything from TTS to translations. Every session can also be exposed through built-in OpenAI- and Anthropic-compatible API servers, so other applications can use your local model as if it were a cloud endpoint. Everything runs on your own machine — 100% private, no account, no telemetry — with one-click installers for Windows, Linux and macOS. The project, maintained by oobabooga and contributors, is licensed under AGPL-3.0 and remains the default front end for the local-LLM community.

Key Features

• Multiple inference backends – llama.cpp, Transformers, ExLlamaV2 and more, switchable per model.
• Chat, notebook and API modes – one app for conversation, raw completion and programmatic access.
• Vision and tool calling – multimodal models and function-calling workflows supported.
• OpenAI-compatible server – serve your local model to any compatible app or library.
• Extensions ecosystem – add speech, translation, image generation and other capabilities.
• LoRA training tools – fine-tune adapters on your own data from the interface.
• One-click installers – get running on Windows, Linux or macOS without manual setup.

💳 Pricing

Text Generation Web UI is free and open source under the AGPL-3.0 license. There is no paid version; your only cost is the hardware the models run on.

Who Is It For

Local-LLM enthusiasts, writers, roleplayers and developers who want a complete, private front end for open models — from casual chatting to serving an API for their own projects.