Anthropic发布Claude Haiku 5.5：对标GPT-6
Claude Haiku 5.5
Haiku系列是开发者高频调用的主力模型，此次直接对标竞品定价并调整计费策略，对Agent开发者的成本控制有直接影响，建议关注其实际调用表现。
As previously promised, here's Anthropic's new fast, low cost model: Introducing Claude Haiku 5.5.
如先前承诺，以下是 Anthropic 的新款快速、低成本模型：介绍 Claude Haiku 5.5。
The previous Haiku, 4.5, was very much showing its age. It came out almost a year ago, and was priced at $1/million input and $5/million output - relatively expensive even back then, and a full 10x the price of OpenAI's GPT-6 Luna, released last month.
上一代 Haiku 4.5 已显老态。它发布于近一年前，定价为每百万输入 token 1 美元、每百万输出 token 5 美元——即便在当时也算相对昂贵，且价格是上个月发布的 OpenAI GPT-6 Luna 的整整 10 倍。
The new Haiku exactly matches the price of GPT-6 Luna - $0.10/$0.50 - up to 100,000 tokens. Beyond 100,000 tokens the price increases 5x to $0.50/$2.50. Luna itself has a price increase at 272,000 tokens but only to $0.20/$0.75.
新款 Haiku 的价格与 GPT-6 Luna 完全一致——100,000 token 以内为 $0.10/$0.50。超过 100,000 token 后，价格上调 5 倍至 $0.50/$2.50。Luna 本身在 272,000 token 处也有涨价，但仅涨至 $0.20/$0.75。
Haiku 5.5 also uses a new, less generous tokenizer. My Claude Token Counter tool shows that the same long prompt uses around 1.25x as many tokens with Haiku 5.5 compared to Haiku 4.5, so there's a hidden price increase there.
Haiku 5.5 还采用了一种新的、不那么慷慨的分词器。我的 Claude Token Counter 工具显示，与 Haiku 4.5 相比，相同的长提示词在 Haiku 5.5 中使用的 token 数量约为 1.25 倍，因此存在隐性涨价。
If your workloads fit in 100,000 tokens, Haiku is the same price as Luna and reports higher benchmark scores. Above 100,000 tokens, Luna looks like a much better deal.
如果你的工作负载在 100,000 token 以内，Haiku 的价格与 Luna 相同，且基准测试得分更高。超过 100,000 token 时，Luna 看起来是更划算的选择。
The most recent release of llm-anthropic finally fixed it so I don't need to ship a new version of that plugin for every new model. I tested the new model like this:
llm-anthropic 的最新版本终于修复了问题，我不再需要为每个新模型发布该插件的新版本。我通过以下方式测试了新模型：
llm install -U llm-anthropic
llm anthropic refresh
llm -m claude-haiku-5.5 "Generate an SVG of a pelican riding a bicycle" -o thinking_effort lowPelicans
鹈鹕
Here are pelicans for low, medium, high, xhigh, and max. The new Haiku doesn't let you disable reasoning, and defaults to medium. I got a good bicycle frame for everything beyond low. The low effort pelican cost 0.0936 cents and took 7 seconds.
以下是低、中、高、超高和最大努力程度下的鹈鹕图像。新款 Haiku 不允许禁用推理过程，并默认使用中等强度。除最低强度外，其他所有强度都生成了不错的自行车车架。最低强度的鹈鹕花费 0.0936 美分，耗时 7 秒。
This max effort pelican took 5 minutes 9 seconds to generate, but still only cost me 3.3826 cents:
这只最大努力程度的鹈鹕生成耗时 5 分 9 秒，但仍仅花费我 3.3826 美分：
For comparison, here's the pelican I got a year ago from Haiku 4.5 (for 0.7583 cents - Haiku 4.5 did not support reasoning levels). It sucked at drawing pelicans:
作为对比，这是我在一年前从 Haiku 4.5 获得的鹈鹕（花费 0.7583 美分——Haiku 4.5 不支持推理强度设置）。它在绘制鹈鹕方面表现糟糕：
And a generous API credit scheme for subscribers
以及面向订阅用户的慷慨 API 积分方案
In addition to Haiku 5.5, Anthropic announced today that they are halving the price of cache reads for Sonnet 5.5. They've also added API credits to subscription plans:
除了 Haiku 5.5，Anthropic 今天还宣布将 Sonnet 5.5 的缓存读取价格减半。他们还在订阅计划中增加了 API 积分：
Second, this week, we’ll roll out a new monthly API credit to all Max and Team subscribers for use on the Claude Platform. Max 5x users will get $100 in credits per month, Max 20x users will get $200, and Team subscribers will receive up to $500, pooled across their users.
其次，本周我们将向所有 Max 和 Team 订阅用户推出新的月度 API 积分，用于在 Claude Platform 上使用。Max 5x 用户每月将获得 100 美元积分，Max 20x 用户将获得 200 美元，Team 订阅用户最多可获得 500 美元积分，在其用户间共享池化。
Claiming this is pleasantly easy: navigate to Settings -> Billing and select the API organization that should benefit from the credits every month:
领取此福利非常方便：导航至设置 -> 账单，然后选择每个月应受益于这些积分的 API 组织：
The API credits exactly match the cost of the subscription itself. This is really generous - it makes it much easier for subscribers to use the API. Anthropic also let you disable auto-reload for the API, with the consequence that "API requests will stop when your balance runs out" - exactly what you want if you're planning to burn through those API credits without risk of a nasty billing surprise.
API 额度与订阅本身的成本完全匹配。这非常慷慨——它让订阅者更容易使用 API。Anthropic 还允许你禁用 API 的自动续订，后果是“当你的余额耗尽时，API 请求将停止”——如果你计划在不担心意外账单惊喜的情况下用完这些 API 额度，这正是你想要的。
Note that the monthly credits do not roll over - use them or lose them.
请注意，月度额度不会累积结转——要么使用它们，要么作废。
OpenAI still allow you to use your Codex subscription for personal API use, which works out as a better deal for heavy API users. This new credit scheme goes at least some way to overcoming that difference.
OpenAI 仍然允许你将 Codex 订阅用于个人 API 使用，这对重度 API 用户来说更划算。这一新的信用额度方案至少在某种程度上弥补了这一差异。
Tags: ai, generative-ai, llms, anthropic, claude, llm-pricing, pelican-riding-a-bicycle, llm-release
标签：ai, generative-ai, llms, anthropic, claude, llm-pricing, pelican-riding-a-bicycle, llm-release
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力