Expired85% confidenceFactTime unknown
Claude Opus 4.5 scored higher than all human candidates ever evaluated by Anthropic in a two-hour timed internal engineering exam.
1
Sources
85%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 8/29/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedIn the SABench benchmark, Claude Opus 4.7 scored approximately 33, while human senior engineers scored 80–90.75% similarUnverifiedClaude Opus 4.6 leads in performance on real-world expert-level tasks in the SWE-bench benchmark.73% similarUnverifiedIn the architect landing page test, Claude Opus 4.5 and Minimax M2.1 tied for the fastest completion time at 1 minute 38 seconds.72% similarUnverifiedClaude Opus 4.8 achieved a score of 1890 in agentic performance benchmarks, 137 points higher than the previous generation and 121 points ahead of GPT 5.5.69% similarUnverifiedClaude Opus 4.8 has reached top-tier performance among its generation across coding, agentic tasks, reasoning, and knowledge work according to Anthropic68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/13793API
curl https://kongchang.com/api/v1/knowledge/claims/13793MCP
get_claim(id=13793)