Archive — US — turtlesai
OpenAI o1-Pro now in API: a more powerful but costly reasoning LLM
OpenAI has launched o1-pro, a powerful reasoning AI model available via API for select developers. Offering improved performance at a high cost ($150/million input tokens, $600/million generated), evaluations show mixed results, with improvements in stability but challenges in specific tasks.
Key points
- Enhanced performance: Utilizes increased processing power for accurate responses.
- Selective access: Limited to developers spending at least $5 on OpenAI services.
- High cost: $150 per million input tokens and $600 per million output tokens.
- Mixed evaluations: Improvements seen, but struggles with complex tasks like puzzles.
Why it matters
- Reflects a trend in AI towards more powerful and versatile models as seen in competitors releasing high-performance systems.
- Demonstrates the rising operational costs of sophisticated AI solutions, a notable trend in tech.
- Aligns with the growing demand for AI models capable of advanced reasoning in various applications.
The introduction of o1-pro signifies OpenAI’s commitment to advancing AI capabilities, albeit with significant costs and some mixed initial feedback that poses challenges for developers.