Definition
text-generation-webui, widely known by its maintainer's handle “oobabooga,” is an open-source Gradio web interface for running large language models locally. It is designed to be fully offline and private, with no telemetry, and it is one of the longest-standing graphical front ends for self-hosted inference. For a sovereign user, it provides a browser-based control panel over models running entirely on their own machine.
Multiple backends in one interface
A defining trait of text-generation-webui is that it does not bind to a single inference engine. It can load models through several backends, including llama.cpp for GGUF models, ExLlamaV2 and ExLlamaV3 for EXL-format models, Hugging Face Transformers, and TensorRT-LLM. The user can switch between backends and models from the interface, which makes it useful as a single place to compare formats and runtimes without reinstalling tooling for each.
Features and access
Beyond chat, it supports text generation, vision-capable models, tool calling, and even local fine-tuning workflows. It exposes an OpenAI- and Anthropic-compatible API, so it can act as a local drop-in endpoint for software written against those hosted APIs. Portable, unzip-and-run builds exist for GGUF models, alongside a one-click installer for the full feature set on Windows, Linux, and macOS.
This UI ties together several engines covered elsewhere in this glossary; see llama.cpp and ExLlamaV2 for two of the backends it can drive.
Compare local-AI front-ends in the sovereign self-hosting catalog.
In Simple Terms
text-generation-webui, widely known by its maintainer’s handle “oobabooga,” is an open-source Gradio web interface for running large language models locally. It is designed to be…