The completions endpoint can be used for a wide variety of tasks - you input some text as a prompt, and the model will generate a text completion that attempts to match the given context or pattern.
For example, if you give the API the prompt, “As Shakespeare said, “All the world's a stage, it'll return “and all the men and women merely players.” with high probability.
This means you can "program" the model by providing instructions or just a few examples of what you'd like it to do. Its success generally depends on the complexity of the task and the quality of your prompt.
See prompt engineering to guide you through this.
While all prompts result in completions, text completion can be seen as its own task in instances where you want the model to pick up a paragraph or phrase where you left off.
Updated about 2 months ago