Unverified50% confidenceBenchmarkExact time
On the 2024 AIME math exam, OpenAI o1 averaged 70%, scored 74% on a single attempt, and reached 83% after 64-sample voting, while GPT-4o scored only 12%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/30/2026
First Seen
Valid until: 1/28/2027
Sources
14 Core Concepts of Reasoning Models: From Chain of Thought to GRPO Explained
youtubeBrix AI10/29/2026
Related Entities
Related Claims
UnverifiedThe final version of DeepSeek R1 and the December 2024 version of OpenAI o1 trade wins on benchmarks: 79.8% vs 79.2% on AIME 2024, 97.3% vs 96.4% on Math500, with o1 slightly ahead on Codeforces.71% similarUnverifiedIn a certain programming benchmark, OpenAI's GPT-5.5 scored 76% in the benchmark's built-in Harness, but only 57% within Codex internally.68% similarUnverifiedThe mathematical results released by OpenAI include 719 top-level results, of which 300 were formalized in Lean, about 42%.67% similarUnverifiedOn the SimpleQA benchmark, o4-mini declines to answer only 1% of questions, with an error rate of 75%.67% similarUnverifiedOpenAI announced that Astra achieved 98% on Frontier Math Tier 4.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1116033API
curl https://kongchang.com/api/v1/knowledge/claims/1116033MCP
get_claim(id=1116033)