Unverified65% confidenceFactTime unknown
在独立测试者的综合排名中，Claude 4.5 Haiku分别排在第34和第37位，是主流AI实验室新一代模型中得分最低的一个
1
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026
Sources
Claude 4.5 Haiku实测翻车：编码能力全面溃败，性价比被竞品碾压
bilibili烟神殿殿主
Related Entities
Related Claims
UnverifiedClaude Haiku 4.5在智能体工具使用的零售基准测试中得分83.2%，航空任务得分63.6%，电信任务得分83%73% similarUnverifiedClaude Haiku 4.5在三个前端测试场景中生成速度最快，成本最低73% similarUnverified独立测试者对Claude 4.5 Haiku的测试反复运行了超过五次，结果始终表现不佳71% similarUnverifiedClaude Haiku 4.5的错误拒绝率低于Sonnet 4.5和Opus 4.1，是Anthropic迄今最安全的模型之一69% similarUnverifiedHaiku 4.5的代理工具使用评分达83%，而Claude 4仅为9.6%67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/24352API
curl https://kongchang.com/api/v1/knowledge/claims/24352MCP
get_claim(id=24352)