Expired65% confidenceFactExact time
Claude Code may default to calling a Sonnet-level model rather than the highest-level Opus 4.
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Related Claims
UnverifiedClaude Code can point sub-agent models to lightweight versions (such as Claude Haiku or Sonnet) while the primary agent uses a high-performance version (such as Claude Opus or Sonnet 4) to control costs77% similarPartially VerifiedIn code generation scenarios, the Claude Sonnet series is currently considered the best-performing among Cursor's built-in models.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/45558API
curl https://kongchang.com/api/v1/knowledge/claims/45558MCP
get_claim(id=45558)