Expired60% confidenceFactTime unknown
Haiku 4.5 reached 90% of Sonnet 4.5's performance on the Agentic Coding evaluation.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedClaude Haiku 4.5 can already match the previous generation flagship model Sonnet 4 in programming performance80% similarUnverifiedHaiku 4.5 has surpassed all competing models in the same tier in the field of agentic coding.72% similarUnverifiedClaude Haiku 4.5 surpassed Claude Sonnet 4 in Computer Use capability.72% similarUnverifiedClaude Sonnet 4.5 significantly outperforms the previous-generation Sonnet 4 model on software engineering benchmarks, and even surpasses the higher-positioned Opus model.66% similarUnverifiedIn actual programming tests, for the weather card task, Haiku 4.5 produced a fully functional result with animation effects, while Sonnet 4.5 had animation bugs, and Opus 4.1 performed the best.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/24436API
curl https://kongchang.com/api/v1/knowledge/claims/24436MCP
get_claim(id=24436)