What is TextGen used for?

TextGen is an open-source desktop app for running large language models locally, with chat, vision, tool-calling and an OpenAI-compatible API. It is a good fit for private local chat with open models and local drop-in API for OpenAI-style clients.

Is TextGen open source?

Yes. TextGen is open source under the AGPL-3.0 licence. Its source code is on GitHub at oobabooga/textgen and is written mainly in Python.

Yes. TextGen is open source, so the software itself is free to use.

Can I self-host TextGen?

Yes. TextGen can be self-hosted on your own server or infrastructure; there is no official hosted version.

What is TextGen an alternative to?

TextGen is an open-source alternative to ChatGPT and LM Studio. Other open-source alternatives to ChatGPT include Jan, AnythingLLM and Open WebUI.

Is TextGen actively maintained?

Yes. The most recent commit to TextGen was on 17 August 2026, and the latest release is v4.9, published on 20 May 2026. The project has 48k stars on GitHub.

TextGen: open-source alternative to ChatGPT, LM Studio

About TextGen

TextGen is a desktop application for running large language models on your own machine. It provides a chat interface and an API, collects no telemetry, and keeps conversations on the local computer. Portable builds for Linux, Windows and macOS include all dependencies and come in CUDA, Vulkan, ROCm and CPU-only variants, so a user can download, unzip and launch it.

Chat modes include instruction-following, chat-instruct and character chat, with prompts formatted by Jinja2 templates. The app handles vision input, file attachments such as PDFs and .docx documents, message editing and conversation branching, plus a notebook tab for free-form generation. It supports several backends, including llama.cpp, Transformers, ExLlamaV3 and TensorRT-LLM, and models can be switched without restarting.

An OpenAI- and Anthropic-compatible API with tool calling lets it stand in locally for those services. Tools are single Python files, and MCP servers are supported. Additional features include LoRA fine-tuning and image generation. TextGen is released under the AGPL-3.0 license.

Key features

Runs large language models locally
Vision input and file attachments
OpenAI- and Anthropic-compatible API
Tool calling with Python tools and MCP
Multiple backends including llama.cpp
LoRA fine-tuning and image generation

Good fit for

→Private local chat with open models
→Local drop-in API for OpenAI-style clients

Built with: Python
Tags: llm; local-ai; chat; openai-compatible-api; llama-cpp; desktop; privacy; fine-tuning

TextGen: questions and answers

What is TextGen used for?: TextGen is an open-source desktop app for running large language models locally, with chat, vision, tool-calling and an OpenAI-compatible API. It is a good fit for private local chat with open models and local drop-in API for OpenAI-style clients.
Is TextGen open source?: Yes. TextGen is open source under the AGPL-3.0 licence. Its source code is on GitHub at oobabooga/textgen and is written mainly in Python.
Is TextGen free?: Yes. TextGen is open source, so the software itself is free to use.
Can I self-host TextGen?: Yes. TextGen can be self-hosted on your own server or infrastructure; there is no official hosted version.
What is TextGen an alternative to?: TextGen is an open-source alternative to ChatGPT and LM Studio. Other open-source alternatives to ChatGPT include Jan, AnythingLLM and Open WebUI.
Is TextGen actively maintained?: Yes. The most recent commit to TextGen was on 17 August 2026, and the latest release is v4.9, published on 20 May 2026. The project has 48k stars on GitHub.