Unverified50% confidenceBenchmarkExact time
Claude 3.5 Sonnet在HumanEval基准测试上的得分超过90%，而2021年最初的Codex模型得分仅为28.8%
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
Vibe Coding入门：零基础用自然语言与AI一起写软件
Related Claims
VerifiedClaude 3.5 Sonnet achieved a 92% pass rate on the code generation benchmark HumanEval.79% similarVerifiedClaude 3.7 Sonnet scored above 60% on the SWE-bench Verified benchmark.77% similarUnverifiedClaude 3.5 Sonnet surpassed a 49% solve rate on the SWE-bench software engineering benchmark.72% similarUnverifiedClaude Sonnet 4.6 scored higher than Opus 4.6 on the GDPVal AA benchmark.72% similarVerifiedClaude Sonnet 5 scores 63.22% on SWE-bench Verified (Agentic Coding), approximately 6 percentage points behind Opus 4.8.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/503708API
curl https://kongchang.com/api/v1/knowledge/claims/503708MCP
get_claim(id=503708)