OpenAI Launches o1-Pro API, Its Most Expensive AI Model Yet

The o1-pro model offers advanced reasoning capabilities but faces scrutiny over steep pricing and marginal performance gains.

The o1-pro API is now available to select developers on OpenAI's paid tiers, priced at $150 per million input tokens and $600 per million output tokens.
The model features a 200,000-token context window and supports text and image inputs but lacks audio input capabilities.
ChatGPT Pro subscribers have had access to o1-pro since December 2024, with rate limits but no additional costs.
OpenAI claims the model offers more reliable responses and advanced features like vision support, function calling, and structured outputs.
Developer feedback and internal benchmarks suggest only slight performance improvements over the standard o1 model, raising questions about its value.