Max token output for GPT-4 (Non-Turbo)?

GPT-4-Turbo has an output limit of 4095, but does GPT-4 (non-turbo) have an output limit? I couldn’t find this information anywhere.

1 Like

Playground is showing 8k, but I’m not sure… It’s not listed on the models page anymore…


There is no artificial limit on response. You can specify a high max_tokens, or leave it out of the API call to potentially exhaust the entire context length.