# oobabooga/textgen

Open-source desktop app for local LLMs. Text, vision, tool-calling, OpenAI/Anthropic-compatible API. 100% private.

Repository: https://github.com/oobabooga/textgen
Canonical: https://ross.abutalabs.com/products/textgen
Language: Python
License: AGPL-3.0
License Family: copyleft
Last push: 2026-08-17T17:38:16+00:00

## Health v2 (maintenance only)
Score: 88/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 98, release rhythm 72, longevity 96
- inputs: {"age_days": 1351, "days_push": 16, "days_rel": 105, "gap_med": 7.0, "n_releases_24m": 63}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 47585, forks 5983 (observed 2026-08-28T04:12:11.262631+00:00)

## What it is
TextGen is an open-source desktop application for running large language models locally with full privacy and no telemetry. It provides a chat UI with vision, file attachments, and tool-calling, plus an OpenAI/Anthropic-compatible API server backed by multiple inference engines like llama.cpp, ExLlamaV3, and Transformers.

## Use cases
- run llm locally on my own gpu
- chat with gguf models offline
- local openai api replacement
- talk to pdf documents with a local model
- fine-tune a lora on chat data
- generate images with diffusers locally
- use mcp tools with a local llm
- private chatgpt alternative without telemetry

## When to choose
- you want a one-click desktop app for local LLM inference with no setup
- you need an OpenAI/Anthropic-compatible local API for development
- privacy matters and you need fully offline chat with vision and file support
- you want to switch between multiple backends (llama.cpp, ExLlamaV3, Transformers) without restarting

## When to avoid
- you need a headless inference server for production clusters rather than a desktop app
- you only want a lightweight CLI or programmatic library without a GUI
- you require multi-user hosted deployment with access control

## Facets
- artifact type: application
- maturity: active
- function: llm-inference, chat-interface, http-server, rag, prompt-engineering, mcp, stable-diffusion, llm-training, gui
- domain: large-language-models, artificial-intelligence, self-hosted, privacy, developer-tools
- platform: windows, python, self-hosted
- tags: local-llm, gguf, llama-cpp, openai-compatible-api, text-generation-webui, character-chat, lora-finetuning, vision-models, tool-calling, offline-ai, linux, macos, desktop, gpu

## Member repositories
- oobabooga/textgen (main) score 88

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:11.262631+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:21:00.991522+00:00, confidence not recorded.
  - readme: https://github.com/oobabooga/textgen (fetched 2026-08-28T04:12:11.262631+00:00, sha bff8a50d6016)
- Data as of 2026-08-30T08:39:29.467469+00:00.