Meta 发布其最小模型 Muse Ember，称其智能水平与 Claude Haiku 5.5 相当，成本约为后者的三分之一。该模型面向智能体全天重复的工具调用、编码子任务、路由、信息抽取和长文档解析。定价为 200K tokens 以内每 1M tokens 输入 $0.10、输出 $0.50，超过 200K tokens 为输入 $0.30、输出 $1.50；作者指出 Haiku 在 100K tokens 以下费率相同，Ember 的成本优势主要来自每个任务消耗更少 token。
同一件事，精选收录的是《Meta 发布最小最高效模型 Muse Ember》
Meta released Muse Ember, its smallest model, which it says matches Claude Haiku 5.5’s intelligence for about a third of the cost.
aimed at the tool calls, coding subtasks, routing and extraction agents repeat all day.
up to 200K tokens: $0.10 input, $0.50 output per 1M tokens
over 200K tokens: $0.30 input, $1.50 output per 1M tokens
Haiku charges identical rates for prompts under 100K tokens, so Ember’s cost edge comes mainly from spending fewer tokens per task.
Releasing Muse Ember today -- our smallest model that delivers higher intelligence per dollar than others in its class. Built for token-heavy work that agents repeat all day: tool calls, coding subtasks, routing, extraction, and long document parsing. Available today on the Meta Model API and in Muse Code.在 X 查看被引用的帖子
来源：Rohan Paul · x.com