Meta 发布 Muse Ember：36 分追平 Claude Haiku 5.5，成本仅三分之一
Meta 这次的思路很明确：知识面窄没关系，把幻觉率砍半、把成本压到 $0.14 一任务，专攻自动化执行。在 Terminal-Bench 上它反超 Haiku 5.5，自家的 Muse Code 里更是领先 Claude Code 和 Codex。跑批量任务的团队可以重点看看。
Meta 推出高效档新模型线 Muse Ember，在 Artificial Analysis Intelligence Index 上得 36 分，与 Claude Haiku 5.5 (max) 持平，但 Cost per Task 仅 $0.14，约为后者的三分之一。xhigh 档得 34 分，成本 $0.12。定价为每百万 tokens 输入 $0.10、缓存输入 $0.004、输出 $0.50，与 Claude Haiku 5.5 和 GPT-6 Luna 相同。
技术上，Muse Ember (max) 在 Terminal-Bench 4.0 上得 40.9%，超过 Claude Haiku 5.5 (max) 的 32.8%，但在 FrontierCode 2.0 上以 27.2% 落后于对方的 34.9%，智能体编程表现两极。在 AutomationBench-AA 上得 68.1%，与 Muse Wonder (max) 持平；GDP.pdf 子项 22.8% 高于 Haiku 5.5 的 20.8%，AA-Briefcase 则以 1521 Elo 落后于 1574。在自家 Muse Code 智能体中，其 Coding Agent Index 得 46 分，高于 Claude Code 中的 Haiku 5.5（37）和 Codex 中的 GPT-6 Luna（31）。
知识覆盖是主要短板：AA-Omniscience 准确率 21.1%，低于 Haiku 5.5 (max) 的 34.7% 和 Gemini 3.8 Flash (high) 的 50.5%，主因是拒答 64% 的问题，幻觉率因此压到 17.4%，约为 Haiku 5.5 的一半。与同门 Muse Wonder (max) 相比，它落后 8 分，但成本仅为后者的约 1/50（$0.14 vs $6.95），走的是低成本自动化执行路线。
Meta's newest model Muse Ember scores 36 on the Artificial Analysis Intelligence Index, matching Claude Haiku 5.5 (max) at a third of the Cost per Task
Muse Ember is Meta's newest efficient-tier model line, with the same input and output token prices as Claude Haiku 5.5 and GPT-6 Luna
Key Takeaways:
➤ On the Intelligence Index vs. Cost per Task Pareto frontier at both max and xhigh effort. Muse Ember (max) is the cheapest model scoring 36 or higher, at $0.14 per task, and Muse Ember (xhigh) scores 34 for $0.12. Claude Haiku 5.5 (max) costs almost 3x more at $0.41 for the same score as Muse Ember (max). Muse Ember (max) is also about 4x cheaper than DeepSeek V4.1 Flash 1017 (max, 33, $0.60) and GLM-5.3-Flash (28, $0.58)
➤ 8 points behind Muse Wonder for about 1/50th of the Cost per Task. Muse Wonder (max) scores 44 at $6.95 per Intelligence Index task
➤ Stands out on AutomationBench-AA. Muse Ember (max) scores 68.1% on AutomationBench-AA, level with Muse Wonder (max), and beats Claude Haiku 5.5 (max) on GDP.pdf (22.8% vs 20.8%), but trails it on AA-Briefcase (1521 vs 1574 Elo)
➤ Mixed in agentic coding in our Intelligence Index evaluations. It beats Claude Haiku 5.5 (max) on Terminal-Bench 4.0 (40.9% vs 32.8%) but trails it on FrontierCode 2.0 (27.2% vs 34.9%)
➤ Ahead of similarly priced models when running in its own coding agent. In Muse Code, Muse Ember (max) scores 46 on the Artificial Analysis Coding Agent Index, ahead of Claude Haiku 5.5 in Claude Code (max, 37) and GPT-6 Luna in Codex (max, 31)
➤ Knowledge is the main weakness, but it rarely hallucinates. Its AA-Omniscience accuracy of 21.1% is the lowest of the similarly intelligent models we compared, against 34.7% for Claude Haiku 5.5 (max) and 50.5% for Gemini 3.8 Flash (high). This is mostly because it declines to answer 64% of questions, which keeps its hallucination rate at 17.4%, half of Claude Haiku 5.5’s (max)
Other details: ➤ Pricing: $0.10 input / $0.004 cached input / $0.50 output per 1M tokens ➤ Reasoning effort: results shown at max effort, with xhigh also on the Intelligence vs. Cost chart ➤ Availability: Meta's first-party API and Muse Code
This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Muse Ember, Meta, Claude Haiku 5.5, GPT-6 Luna.
Source attribution: curated from Artificial Analysis (@ArtificialAnlys).
