Model detail
o1-2024-12-17
The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding. The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought. The o1 models are optimized for math, science, programming, and other STEM-related tasks. They consistently exhibit PhD-level accuracy on benchmarks in physics, chemistry, and biology. Learn more in the [launch announcement](https://openai.com/o1).
Model specs
- Context length
- 200K
- Max output
- 100K
- I/O modalities
- Text / Image
- Released
- 2024-12
02
API endpoints
One MixRoute gateway, OpenAI-compatible
-
OpenAI-compatible
/v1/chat/completionsPOST
03
Pricing
Flat rate, unit: /1M Tokens
Input
$15.0000 /1M Tokens
Completion
$60.0000 /1M Tokens
Cache read
$7.5000 /1M Tokens
05
Selection summary
Quickly judge whether it fits your workload
What is this model good for?
PhD-level STEM reasoning with chain-of-thought
Use o1 for math, science, programming, and STEM tasks requiring deep reasoning. It uses chain-of-thought to think before answering, achieving PhD-level accuracy on physics, chemistry, and biology benchmarks.
What should you check before using it?
Plan for longer thinking time and higher cost
o1 spends more time thinking before responding, which increases latency and cost. Test your STEM tasks to confirm the accuracy improvement justifies the additional compute time.
Why use it through MixRoute?
Use the compatible endpoint with stable model ID
Use the confirmed compatible endpoint with model ID o1 through MixRoute, then run the same integration tests used for the current client.
08
Token cost estimator
Live estimate from this page's pricing, not an actual bill
Share of the same prefix read repeatedly, up to 100%
09
FAQ
o1 is OpenAI’s reasoning model trained with large-scale reinforcement learning to think before responding using chain of thought. It is optimized for math, science, programming, and STEM tasks, consistently achieving PhD-level accuracy on physics, chemistry, and biology benchmarks.
One endpoint, a testable decision
Test this model and alternate routes with the same request format
Start from a real workload, then let quality, total cost, and failure conditions decide whether to send production traffic.