Expired65% confidenceFactTime unknown
The Sonnet model underlying Claude Code consistently leads on mainstream programming benchmarks such as HumanEval and SWE-bench.
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedOpenAI's o-series models and Anthropic's Claude 3.5 Sonnet have achieved scores on code generation benchmarks (such as HumanEval and SWE-bench) that approach or exceed the average level of human programmers.77% similarUnverifiedIn the Pillow library RGB-to-HSV precision bug test, Claude Code using the Sonnet 4.6 model passed on the first attempt with clear logic, earning a perfect score.76% similarVerifiedClaude models have consistently led on programming capability benchmarks such as SWE-bench.72% similarUnverifiedThe tester proposed that the optimal workflow is for Sonnet to handle planning, Haiku sub-agents to execute specific operations, and Sonnet to handle final review.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/17488API
curl https://kongchang.com/api/v1/knowledge/claims/17488MCP
get_claim(id=17488)