ARC Prize tests o3 and o4-mini on ARC-AGI
Original title:Analyzing o3 and o4-mini with ARC-AGI
AI summary
The ARC Prize Foundation published the first public test results for OpenAI o3 and o4-mini on ARC-AGI. o3-low scored 41% on the ARC-AGI-1 Semi Private Eval and o3-medium reached 53%, while neither exceeded 3% on ARC-AGI-2; o4-mini-low scored 21% on ARC-AGI-1.
Key Takeaways
- The ARC Prize Foundation published the first public test results for OpenAI o3 and o4-mini on…
- o3-low scored 41% on the ARC-AGI-1 Semi Private Eval and o3-medium reached 53%,while neither…
- o4-mini-low scored 21% on ARC-AGI-1.
Quick Facts
- Published
- Updated
- Source
- ARC Prize:官方博客
- Category
- Tutorials & Tips
- AI score
- 70
ARC Prize 公布了自家基准上 o3 与 o4-mini 的完整实测数据和高推理档的失真问题,读者可借此校准对推理档位选择的预期。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: ARC Prize tests o3 and o4-mini on ARC-AGI. April 22, 2025. https://new.nat6.net/en/news/cmtyo2p2s03g6rog05alsxq1q/