Anthropic 发布 Claude Haiku 5.5，价格最高降 90%
Claude Haiku 5.5 arrives with massive price cuts proving the AI pricing arms race is far from over
Anthropic 发布 Claude Haiku 5.5，这是其目前最快、最便宜的小模型，平均价格比 Haiku 4.5 低约 75%，10 万 token 以内提示词最高降价 90%，超过 10 万 token 则贵 5 倍。基准测试显示其计算机使用能力提升明显，OSWorld-2.1 得分从 15.7% 升至 72.4%，Terminal-Bench 4.0 从 0 升至 39.2%，Humanity's Last Exam 无工具 45.9%、有工具 57.4%。该模型是首个支持可调推理等级的 Haiku 级模型，已在 AWS、Google Cloud 和 Microsoft Azure 上线，同时 Sonnet 5.5 的缓存读取费用从每百万 token 0.20 美元降至 0.10 美元。
推荐理由
精选AI 评分 84
Haiku 5.5 的降价幅度与计算机使用能力跃升同时出现，读者可据此判断小模型在智能体任务中的成本边界。
主题
正文
当前显示原文，中文全文译文尚未就绪。
Key Points
Anthropic has released Claude Haiku 5.5, the company's fastest and most affordable small model to date. Benchmark results show a major performance jump over its predecessor, and Anthropic is also cutting prices for Sonnet 5.5.
Haiku 5.5 is designed for high-volume, cost-sensitive tasks like summarization, database queries, classification, and live customer support, according to Anthropic. On average, the model costs about 75 percent less than Haiku 4.5. For requests with prompts up to 100,000 tokens, which Anthropic says account for roughly 90 percent of all previous Haiku requests, prices drop by up to 90 percent. Prompts longer than 100,000 tokens cost five times as much.
Anthropic points out that Haiku 5.5 uses an updated tokenizer that consumes slightly more tokens per task than its predecessor. The same thing happened with the Opus 4.x models , where token usage jumped about 30 percent from the tokenizer change alone. Real-world savings are likely smaller than the per-token prices suggest.
Benchmarks show a big leap over Haiku 4.5
Haiku 5.5 scores 1,620 on the knowledge benchmark GDPval-AA v2.1, more than double the 735 its predecessor managed. On Humanity's Last Exam, it hits 45.9 percent without tools and 57.4 percent with tools, up from 10.2 and 18.7 percent.
The biggest jump is in computer use , where the model operates a computer on its own. Since computer use burns through large amounts of tokens, a cheap, fast model like Haiku 5.5 is a natural fit. It scores 72.4 percent on OSWorld-2.1, up from 15.7 percent. On the agentic coding benchmark Terminal-Bench 4.0, it reaches 39.2 percent while Haiku 4.5 scored zero.
Anthropic also lists OpenAI's budget model GPT-6 Luna as a comparison, and Haiku 5.5 leads across every tested category. Sonnet 5.5 reference scores show, however, that Haiku 5.5 still falls well behind Anthropic's larger model.
First Haiku model lets users trade cost for performance
Haiku 5.5 is the first Haiku-class model with adjustable reasoning levels, letting users balance cost against quality. Anthropic says the model works best for narrowly scoped tasks like compaction, summarization, or sub-agent work. For complex agentic coding, Sonnet 5.5 and Opus 5.5 remain the better picks.
Cybersecurity safeguards are tighter than on the predecessor but allow a broader range of defensive tasks than Sonnet 5.5, partly because the model is less capable overall. Penetration testing stays blocked. Organizations with broader needs can apply for Anthropic's verification programs for life sciences and cybersecurity .
Haiku 5.5 is available now across all platforms, including Amazon Web Services, Google Cloud, and Microsoft Azure.
Sonnet 5.5 gets cheaper too, and Anthropic hands out API credits
Alongside the Haiku launch, Anthropic is cutting cache read costs for Sonnet 5.5 by 50 percent, from $0.20 to $0.10 per million tokens. The company says this should reduce costs for most agentic tasks by about 20 percent. The move is almost certainly a reaction to OpenAI's new GPT-6.1 series , showing that AI price wars are being fought harder than ever.
The company is updating its Python and TypeScript SDKs as well, adding beta support for computer use and browser use .