Archive — US — techcrunch
OpenAI’s o1-pro is the company’s most expensive AI model yet
OpenAI has introduced o1-pro, its most expensive AI model, through its developer API. Priced at $150 per million input tokens and $600 per million output tokens, it offers improved performance over previous models. Despite mixed early feedback regarding its reliability in tasks, OpenAI aims to meet developer demand for better AI responses.
Key points
- OpenAI has launched o1-pro, a more powerful AI model using more computing resources.
- o1-pro is only accessible to developers who have spent at least $5 on OpenAI API services.
- The pricing for o1-pro is significantly higher than previous models: $150 per million tokens for input and $600 for output.
- Early feedback noted o1-pro struggled with tasks like Sudoku puzzles and optical illusions.
- Internal benchmarks show o1-pro slightly outperformed the standard o1, offering more reliable answers, particularly in coding and math.
Why it matters
- The launch of o1-pro reflects the ongoing trend in AI where companies prioritize performance at a premium cost.
- The rising prices of AI models might influence developer accessibility and shape AI utilization in various sectors.
- This model's performance issues may signal potential challenges in the evolving landscape of AI, which emphasizes practical reliability.
OpenAI's o1-pro represents a significant investment for developers seeking enhanced AI capabilities, though its mixed initial performance may create hesitation about the high costs involved.