Unverified70% confidenceFactTime unknown
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
1
Sources
70%
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5个编程任务测试中全部失败80% similarUnverifiedClaude Haiku 4.5在鞭炮连锁爆炸测试中6次测试均未能实现连锁爆炸效果79% similarUnverified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一66% similarUnverified在实际性能测试中，GLM 4.6的表现远远优于Claude 4.5 Haiku66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/24361API
curl https://kongchang.com/api/v1/knowledge/claims/24361MCP
get_claim(id=24361)