Expired70% confidenceFactTime unknown
Claude Haiku 4.5 achieved a score of 73.3% on the SWE-bench benchmark.
4
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Claude Haiku 4.5 Review: Sonnet 4-Level Coding at 1% of the Cost
bilibili攒钱换房车的福叔
Related Entities
Related Claims
UnverifiedHaiku 4.5's agentic tool use score reached 83%, while Claude 4 scored only 9.6%78% similarUnverifiedHaiku 4.5 scored 73.3% on SWE Bench, slightly higher than Sonnet 4 and approaching the level of GPT-5 Codex.77% similarUnverifiedClaude Haiku 4.5 scored 73% on graduate-level reasoning tasks, 83% on multilingual question-answering tasks, and 73.2% on visual reasoning tasks.76% similarUnverifiedOn the SWE Bench Verified leaderboard, Qwen3.6 35B achieved a score of 73.4%.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/24363API
curl https://kongchang.com/api/v1/knowledge/claims/24363MCP
get_claim(id=24363)