OpenAI has launched a new API called the Assistants API, which allows developers to build agent-like experiences within their apps. This API enables developers to create assistants with specific instructions that can leverage outside knowledge and utilize OpenAI’s generative AI models and tools. The use cases for this API range from natural language-based data analysis apps to coding assistants and AI-powered vacation planners. The Assistants API is powered by Code Interpreter, a tool that writes and runs Python code in a sandboxed execution environment. Code Interpreter can generate graphs, process files, and solve code and math problems iteratively. Additionally, the Assistants API includes a retrieval component that can augment dev-created assistants with external knowledge and supports function calling to incorporate programming functions. The API is currently in beta and available to all developers, with billing based on per-token rates. OpenAI plans to allow customers to provide their own assistant-driving tools in the future.
