How to Install Text Generation WebUI 2026

Text Generation WebUI (by oobabooga) is a popular open-source web interface for running large language models locally. It supports GGUF models via llama.cpp, Hugging Face Transformers, ExLlamaV2/V3 and more, with a chat interface, an OpenAI-compatible API and LoRA training.

Table of Contents

Requirements

	Minimum	Recommended
GPU	None (CPU with GGUF) or NVIDIA 8 GB	NVIDIA RTX 16–24 GB
RAM	16 GB	32 GB+
Disk	20 GB+	SSD with space for models (5–50 GB each)

How to Install Text Generation WebUI

Option 1: Portable build (easiest)

Download a portable release for your OS from the GitHub releases page, extract it and run it. Portable builds use llama.cpp with GGUF models and need no installation.

Option 2: One-click installer

git clone https://github.com/oobabooga/text-generation-webui
cd text-generation-webui
# Windows
start_windows.bat
# Linux
./start_linux.sh
# macOS
./start_macos.sh

The script installs Miniconda, PyTorch and all dependencies, and asks which GPU you have. When it finishes, open http://127.0.0.1:7860.

Load a Model

Open the Model tab.
Paste a Hugging Face repo, for example a GGUF build of Qwen or Gemma 4, and click Download.
Select the model, choose the loader (llama.cpp for GGUF) and click Load.
Go to the Chat tab and start talking.

Enable the API

Start with --api to get an OpenAI-compatible endpoint on port 5000, so other AI apps can use your local model.

Troubleshooting

Model too slow: use a smaller quantization (Q4_K_M) or offload more layers to the GPU.
CUDA out of memory: lower context length or pick a smaller model.
Installer errors: delete the installer_files folder and run the start script again.

Run It on a GPU Server

No powerful GPU at home? You can run Text Generation WebUI on a rented GPU server instead. RunPod and Vast.ai have text-generation-webui templates. For a fixed monthly price, GPU Mart (RTX A4000 VPS from $119/month, RTX 4090 $409/month) and HOSTKEY (RTX 4090 from €279/month) offer dedicated servers you can keep running 24/7.

FAQ

Do I need a GPU for Text Generation WebUI?

No, GGUF models run on CPU, but a GPU is much faster.

Which models does it support?

Most open LLMs: Llama, Qwen, Gemma, Mistral, DeepSeek distills and more, in GGUF, safetensors or EXL2/EXL3 formats.

Conclusion

Text Generation WebUI is a powerful all-in-one interface for local LLMs. See also how to run Oobabooga and LM Studio.