In order to gain access to the gpt-4-0314 Model API, you must sign up for the waitlist. The API utilizes the ChatCompletions API, which is also used in gpt-3.5-turbo.
OpenAI plans to invite developers gradually to balance capacity with demand. Additionally, researchers studying the societal impact of AI or AI alignment issues can apply for subsidized access through the Researcher Access Program.
Once access is granted, you will be able to make text-only requests to the GPT-4 model, with image inputs still in limited alpha. OpenAI plans to automatically update the model to the recommended stable version as new versions are released.
Users can also pin the current version by calling gpt-4-0314 , which will be supported until June 14. The pricing structure is $0.03 per 1k prompt tokens and $0.06 per 1k completion tokens, with default rate limits set at 40k tokens per minute and 200 requests per minute.
The GPT-4 model has a context length of 8,192 tokens, but the company is also providing limited access to the 32,768-context version, known as gpt-4-32k. The 32k version will also be automatically updated over time, with the current version being gpt-4-32k-0314 and supported until June 14.
The pricing for the 32k version is $0.06 per 1k prompt tokens and $0.12 per 1k completion tokens. The company is still working on improving the model quality for long context and welcomes feedback on its performance for different use cases. Requests for the 8k and 32k engines are being processed at different rates based on capacity, so users may receive access to them at different times.
Read more articles on the topic:
