Expired50% confidenceBenchmarkExact time
In Matt Mayer's independent benchmark tests, the Opus model scored 77% in Claude Code and 93% in Cursor, with the only variable being the Harness.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/9/2026
First Seen
Valid until: 10/7/2026(expired)
Sources
Related Claims
UnverifiedWhen using Claude Code with the default Opus 4.7 model for investor information extraction tasks, accuracy is typically between 90% and 95%63% similarUnverifiedEven with the most advanced code models, when they claim 'the code has passed testing,' the actual pass rate may be below 70%.63% similarUnverifiedBenchmark overfitting is a core problem in current large model evaluation; some models achieve pass rates above 90% on HumanEval but perform far below expectations in real codebases.62% similarUnverifiedDevin 2.0 resolved 13.86% of real-world programming problems in the SWE-bench benchmark, compared to only 1.96% for previous AI models.61% similarUnverifiedAccording to statistics, Claude Code users choose to approve operation requests 93% of the time, leading to serious approval fatigue.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/400623API
curl https://kongchang.com/api/v1/knowledge/claims/400623MCP
get_claim(id=400623)