o3
Closed
OpenAI's first true "reasoning at scale" model, announced Dec 2024 and publicly released April 2025, which crossed human-expert ceiling on GPQA.
- Publisher
- OpenAI
- Family
- O Series
- Params
- Undisclosed (closed)
- Context
- 200K
- Input
- Output
- Providers
- openai
- License
- Proprietary
- Released
- Apr 2025
Cite
Notes
Only stored in your browser.
Attribution
- Benchmark scores
- AiderAAVaultarena-hard-autoprime-hubfrontiermathosworld-orgSWE-benchOpenRewardVals AIEQ-Bench
Context
200K
Input
$2
/1M tok
Output
$8
/1M tok
Blended
$3.50
#413/ 461/1M · 3:1
Speed
99
#63/ 461tok/s
Cache read
$0.50
0.25x input
Price & speed via Artificial Analysis · ranked across tracked models · caching via OpenRouter
Reported on 29 evals across 12 domains - best 99.2% on MATH-500 · #1 of 190
Reported eval scores
29MATH-50099.2
FrontierMath18.7
Arena-Hard88.8
MMLU-Pro85.3
IFBench71.4
LiveCodeBench80.8
SWE-bench71.7
GPQA Diamond87.7
SciCode41
AIME202491.6
AIME202588.9