Claude Haiku 5.5 在 Code Arena WebDev 首测第 30 名，1587 分
Real-world results are in for Claude Haiku 5.5 (High) by @AnthropicAI.
Arena 公布 Claude Haiku 5.5（High）在 Code Arena: WebDev 的实测结果，以 1587 分首次上榜排在第 30 名，位于帕累托前沿之外但成本效率突出。其价格为每 100 万输入/输出 token 0.10/0.50 美元，与 GPT-6 Luna 持平，得分高出 6 分。相比其他 Claude 模型，Haiku 5.5 比 Haiku 4.5 成本低 90% 且高 257 分，比 Sonnet 5.5（High）低 95% 但低 128 分，比 Opus 5（High）低 98% 但低 70 分，比 Fable 5.1（Max）低 99% 但低 157 分。
推荐理由
精选AI 评分 66
Code Arena 的实测数据给出 Haiku 5.5 在 WebDev 榜单的排名与价格对比，可据此判断小模型在编码场景的性价比位置。
主题
查看原帖Introducing Claude Haiku 5.5: the cheapest, fastest, and most capable small model we’ve ever released. On average, it costs around 75% less to run than Claude Haiku 4.5. https://t.co/jm06cZJfkV
正文
当前显示原文，中文全文译文尚未就绪。
Real-world results are in for Claude Haiku 5.5 (High) by @AnthropicAI.
Debuting at #30 with 1587 pts in Code Arena: WebDev, this release sits just outside the Pareto frontier while still offering strong cost efficiency. It matches GPT‑6 Luna’s price at $0.10/$0.50 per 1M input/output tokens while scoring six pts higher.
Compared with other Claude models, Haiku 5.5 offers: - 90% lower cost than Haiku 4.5 at +257 pts - 95% lower cost than Sonnet 5.5 (High) at -128 pts - 98% lower cost than Opus 5 (High) at -70 pts - 99% lower cost than Fable 5.1 (Max) at -157 pts
Congrats to the team at @AnthropicAI on this release!