Zero-shot prompting
Zero-shot prompting is asking a model to perform a task using instructions alone, without showing it any completed examples.
Zero-shot works well for tasks the model has seen constantly in training — summarising, translating, answering general questions. It works badly when the output must match a specific format or house style, because there is nothing to imitate.
Related terms
- Few-shot prompting
- Few-shot prompting is giving a model several worked examples of a task inside the prompt so it can infer the pattern and apply it to new input.
- One-shot prompting
- One-shot prompting is giving a model exactly one worked example of the task before asking it to perform that task.
More on techniques
See the full glossary, read the guides, or put it into practice in the prompt builder.