There are a few different GPT-4 models to choose from.
One key difference between models is the context length. This is the length of the prompt plus the maximum number of tokens in the completion. Our standard GPT-4 model offers 8,000 tokens for the context. We also offer an extended 32,000 token context-length model, which we are rolling out separately to the 8k model.
Secondly, gpt-4 will refer to our most up-to-date model (and gpt-4-32k for the latest 32k-context model). If you're interested in using a previous snapshot of the model, you can refer to the specific date in the model name, such as gpt-4-0314 or gpt-4-32k-0314. The March 14th snapshot will be available until June 14th.