We recently introduced Chat Completions into the OpenAI API – which currently powers the GPT-4 and ChatGPT APIs. With it, developers now have a powerful new interaction pattern to talk to our models in a similar manner to the ChatGPT experience, as well as the ability to call out to external tools. In this talk, we will be walking through steerability and system prompts, optimizations to improve request latency and throughput with the API, and best practices in deploying production apps on top of our models. Finally, we will also preview how this interface will evolve by covering more recent and nascent capabilities such as plugins and image inputs into our models.
Speaker

Sherwin Wu
Technical Staff @OpenAI
Sherwin is a Member of Technical Staff at OpenAI. He works on the Developer Platform team, which is responsible for the products that allow developers to build products on top of OpenAI models and capabilities.
Speaker

Atty Eleti
Software Engineer @OpenAI
Atty Eleti is a software engineer at OpenAI working on their API and developer platform. Previously, he spent many years building APIs at Stripe. Talk to him about GPT-4, AI, language models, tool use, integrating LLMs into your products, and any feedback on the OpenAI API!