No model is loaded. Please select one in the Model tab.
The message
No model is loaded. Please select one in the Model tab.What it means
You sent a message before any model was in memory. Picking a model in the dropdown only selects it; nothing loads until you press Load, or until a load that failed is fixed.
What to do
Open the Model tab, choose the model and press Load, then wait for “Successfully loaded”. If it shows a traceback instead, that error is the real problem.
You open the web page, type into the chat box, and get this back (the same line also shows up in the terminal, marked ERROR):
No model is loaded. Please select one in the Model tab.Up to version 3.0 it read No model is loaded! Select one in the Model tab., with an exclamation mark, and that's the wording most of the 45 GitHub issues quote (as of September 26, 2026). The project also changed its name along the way. The repository that was oobabooga/text-generation-webui is now oobabooga/textgen, and the README calls the app TextGen.
What it's checking
We read the check in release v4.9 (May 20, 2026). Before generating anything, TextGen looks at whether a model is actually sitting in memory. If none is, you get this message and nothing runs. There are two versions of it. With no model files on disk at all, it tells you to put a GGUF file (the single-file model format llama.cpp uses) into user_data/models first. With files present, you get the "select one in the Model tab" line.
Why it shows up after you picked a model
This is where most people get stuck. In the Model tab, choosing a model from the dropdown doesn't load it. The code behind that dropdown only reads the model's settings and fills in the form, and the status box says Settings updated: Click "Load" to load the model. The model loads when you press the Load button beside it.
The other way to end up here is a load that failed. When Load hits an error, TextGen writes Failed to load the model. to the terminal and puts the Python traceback (the stack of error lines) in the status box. The model slot stays empty, so the next chat message gets "No model is loaded" and the actual error scrolls out of sight. In issue #1255 from 2023 the owner's whole answer was "Select a model in the Model tab", and several people replied that they had. For most of them, a load that had quietly failed is the likely explanation.
How to fix it
Go to the Model tab, pick the model and press Load. Then watch the status box. You want Successfully loaded followed by the model's name, and after that the chat works.
If a traceback appears instead, read its last line. That's the error to search for, and it's a separate problem from this message (this one only ever means the slot is empty).
If the dropdown is empty, the app can't see any model files. Put them in user_data/models inside the TextGen folder (portable builds keep user_data one level up, and --user-data-dir or --model-dir can point elsewhere), then press the refresh button next to the list.
Loading a model every time it starts
The README has a section for this. Add a line like --model my-model.gguf to user_data/CMD_FLAGS.txt, using the name of a file in user_data/models, and the model loads on startup. That also helps if you only use TextGen's API from another program, since the API hits the same check and there's no Model tab to click from there.
Other lines the same feature prints
Match yours against these if the one at the top of the page is not quite it. They come from the same code and mean related things.
No model is loaded! Select one in the Model tab.ERROR No model is loaded! Select one in the Model tab.No model is loaded. Place a GGUF model in your user_data/models folder, then select it in the Model tab.No model is loaded. To get started: 1) Place a GGUF file in your user_data/models folder, 2) Go to the Model tab and select itNo model is loaded