How to download and set up Oobabooga Text Generation Web UI on Windows

This guide demonstrates how to download and set up the Oobabooga Text Generation Web UI on Windows, download a language model from Hugging Face, and test it through the built-in chat interface.

What is Oobabooga Text Generation Web UI?

Oobabooga Text Generation Web UI is an open-source web interface for running large language models locally on your own hardware. It supports a wide range of model formats and backends including llama.cpp, as well as GPU-accelerated inference for NVIDIA and AMD graphics cards. The software is distributed as a portable package directly from its GitHub releases page.

Steps to follow

Download and start the Text Generation Web UI
Download and load an LLM model
Test the model in the chat interface

Example model

# Hugging Face model ID and filename
Qwen/Qwen2.5-1.5B-Instruct-GGUF
qwen2.5-1.5b-instruct-q4_0.gguf

How to download and install Oobabooga Text Generation Web UI video

The following video shows how to download and start the Oobabooga Text Generation Web UI on Windows. The video covers navigating to the GitHub releases page, choosing the correct build, extracting the archive, and running the startup script.

Step 1 - Download and start the Text Generation Web UI

Open your browser and navigate to the Oobabooga Text Generation Web UI releases page on GitHub. The releases page lists all available versions of the software with portable builds for different platforms (Figure 1).

Navigate to releases on Oobabooga GitHub — Figure 1 - Navigate to the releases page on the Oobabooga GitHub repository

Scroll down to the portable builds and choose the one that matches your hardware - select the NVIDIA build if you have an NVIDIA GPU, AMD for AMD graphics cards, or the CPU-only build if you do not have a dedicated GPU (Figure 2).

Choose and download release version — Figure 2 - Choose the appropriate build and download it

Once the download is complete, right-click the ZIP file and select Extract All to unpack the contents to a folder of your choice. The extracted folder contains all files needed to run the application without any additional installation (Figure 3).

Extract the compressed ZIP folder — Figure 3 - Extract the downloaded ZIP archive

Open the extracted folder and double-click start_windows.bat to launch the application (Figure 4).

Run start_windows.bat in the extracted folder — Figure 4 - Run the start_windows.bat script in the extracted folder

A terminal window will appear showing the startup progress. Wait for the server to finish initializing, it will display a local URL when it is ready. Keep this terminal open for the duration of your session (Figure 5).

Wait for Text Generation Web UI to start — Figure 5 - Wait for the Text Generation Web UI server to start

The Text Generation Web UI will open automatically in your default browser, or you can navigate to http://localhost:7860 manually. The dashboard is now ready for you to download and configure a language model (Figure 6).

Figure 6 - The Text Generation Web UI dashboard

Step 2 - Download and load an LLM model

The following video shows how to download a model from Hugging Face, load it in the Text Generation Web UI, and test it through the chat interface.

Click the Model option in the left sidebar to open the model management page. This is where you can download models directly from Hugging Face and manage which model is currently loaded into memory (Figure 7).

Open Models page from the left sidebar — Figure 7 - Open the Models page from the left sidebar

Navigate to the Download model or LoRA section. Enter the Hugging Face model repository ID in the first field and the specific filename in the second field, then click Download. The download will begin immediately and progress will be shown on a progress bar (Figure 8).

Qwen/Qwen2.5-1.5B-Instruct-GGUF
qwen2.5-1.5b-instruct-q4_0.gguf

Enter Hugging Face model ID and start download — Figure 8 - Enter the Hugging Face model ID and filename, then start the download

Once the download completes, click the Refresh button next to the model dropdown to update the list. Select the downloaded model from the dropdown (Figure 9).

Refresh the list and select the downloaded model — Figure 9 - Refresh the model list and select the downloaded model

Click the Load button to load the selected model into memory. The loading process may take a moment depending on the model size and your hardware. A confirmation message will appear in the interface once the model is ready (Figure 10).

Load the model — Figure 10 - Load the selected model

Step 3 - Test the model in the chat interface

Navigate to the Chat tab in the left sidebar to open the chat interface. Type a test message in the input field at the bottom and press Enter or click the send button to submit it to the loaded model (Figure 11).

Open Chat and send a test prompt — Figure 11 - Open the Chat tab and send a test prompt

The model will generate a response and display it in the chat window. A successful reply confirms that the Text Generation Web UI is running correctly and the model has been loaded and is ready for use (Figure 12).

Figure 12 - The LLM response displayed in the chat interface

To sum it up

You have successfully downloaded and set up the Oobabooga Text Generation Web UI on Windows, downloaded a language model from Hugging Face, loaded it into the interface, and verified that it is responding correctly through the chat tab. The application is now ready for use as a local AI chat interface or as a backend for other tools and integrations.