OpenAI o3 is a next-generation reasoning model succeeding o1, with stronger performance on math, science, coding, and multi-step logical problem solving through deeper internal deliberation.
Cost rate
1.5x
Context
200K
Released
Apr 16, 2025
Input
ImageTextFile
Output
Text
Support
ReasoningTool callingStructured outputs
Performance
Median latency and throughput measured across recent requests.
Throughput115 tok/s
Latency5.51s
Best at
The categories where this model ranks highest.
Medicine & Healthcare
#68 · top 17%
Image understanding
#82 · top 49%
Legal & Government
#97 · top 25%
Mathematical
#98 · top 24%
| Domain | Score | Rank |
|---|---|---|
| Software & IT Services | 1455 | #138 |
| Writing, Literature, & Language | 1396 | #138 |
| Life, Physical, & Social Science | 1456 | #104 |
| Entertainment, Sports, & Media | 1394 | #125 |
| Business, Management, & Finance | 1427 | #121 |
| Mathematical | 1448 | #98 |
| Legal & Government | 1452 | #97 |
| Medicine & Healthcare | 1476 | #68 |
| Multimodal | ||
| Image understanding | 1216 | #82 |
| Category | Win rate | Rank |
|---|---|---|
| Data Visualization | 48.1% | #62 |
| Game Development | 56.9% | #126 |
| UI Component | 53.3% | #126 |
| Code Categories | 51.9% | #131 |
| Website | 53.8% | #133 |
Latest news about o3
Recent articles and community posts, linked to their original sources.
OpenAI’s GPT-6.2 Launch Backfired? Real Codex TestSwastik 1,202
OpenAI Just Embarrassed Itself With New GPT 6.2 Astra...AI Revolution31,401
OpenAI Just Dropped GPT-6.2 Astra & Sol?!Julian Goldie SEO2,617
OpenAI GPT-6.2 Astra is Here - It's Not What You ThinkMatt Wolfe120,672
The new ChatGPT can SEE 👀Zayed Almadidy, MD (Dr. Z Neurosurgery)66,487- GPT‑6.2 Astra and SolHacker News1,346862
- GPT-6 CyberHacker News225