Expired50% confidenceFactExact time
On the Diamond subset, GPT-4o achieved a score of 6.3% and a pass rate of 7.2%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Frontier Code Deep Dive: Code That Runs ≠ Code That Merges — A Quality Revolution in Programming Benchmarks
bilibili烟神殿副殿主6/9/2026
Related Claims
UnverifiedOn the Diamond subset, Claude 3.0 Opus scored 5.2%, Gemini 1.5 Pro scored 4.7%, GPT-4o Mini scored 4.6%, and Kimi K2.6 scored 3.8%.79% similarUnverifiedGPT-5.6 set a new all-time high score of 53.6 on the Agent's Last Exam benchmark, surpassing Claude Opus 4.5 by 13.1 points.72% similarUnverifiedIn the Agents' Last Exam benchmark, GPT-5.6 Sol achieved a score of 53.6, which is 13.1 points higher than Claude Fable 5 using adaptive reasoning.71% similarPartially VerifiedClaude Opus 4.8 scored 69.2% on the SWE-bench benchmark, compared to 58.6% for GPT 5.5 and 54.2% for Google Gemini.70% similarUnverifiedOn the Diamond subset, Claude 3.5 Sonnet achieved a score of 13.4% and a pass rate of 14.5%, the best performance among all tested models.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/48342API
curl https://kongchang.com/api/v1/knowledge/claims/48342MCP
get_claim(id=48342)