Meta发布轻量模型Muse Ember，主打Agent工具调用与编码

Today, we're introducing Muse Ember, our smallest and most efficient model designed for complex reasoning execution and the work agents repeat all day: tool calls, coding subtasks, routing, extraction, and long document parsing.

Muse Ember delivers more intelligence per dollar than Claude Haiku 5.5 or GPT-6 Luna at any reasoning setting: it matches Haiku's best score at about a third of the cost and beats every Luna setting. It also sits on the efficiency frontier of the full Artificial Analysis leaderboard: no model scores higher for less per task.

Intelligence vs. Cost Per Task

Artificial Analysis Intelligence Index against cost per task at each reasoning effort, from the raw Artificial Analysis results. Muse Ember costs are estimates at proposed pricing.

Intelligence vs. Cost Per Task$0.00$0.10$0.20$0.30$0.40152025303540Cost Per Task (USD)Artificial Analysis Intelligence IndexxhighmaxmediumhighxhighmaxhighxhighmaxMuse EmberClaude Haiku 5.5GPT-6 Luna

Intelligence vs. Cost Per Task$0.00$0.10$0.20$0.30$0.40152025303540Cost Per Task (USD)Artificial Analysis Intelligence IndexxhighmaxmediumhighxhighmaxhighxhighmaxMuse EmberClaude Haiku 5.5GPT-6 Luna

Artificial Analysis | November 9, 2026. Each point is one reasoning-effort setting. Muse Ember cost at list pricing.

Muse Ember evaluation results

For more details about our evaluations, see our report.

Muse Ember is available today on the Meta Model API, at $0.10 per million input tokens and $0.50 per million output tokens.

In early testing, our partners put Muse Ember to work on the kinds of tasks agents repeat all day, from reviewing contracts to writing code.

Previous slideNext slide

“On the Box AI Complex Work Eval, Muse Ember delivers impressive accuracy, speed and token efficiency. It can score a contract against a company's negotiation playbook clause by clause and land on the right overall risk rating, or screen candidates against a detailed brief and flag the borderline ones who still qualify as alternates. Muse Ember is a fast, efficient foundation for agents doing real work on enterprise content.”

Author

Role

VP of AI Products

Company

“It's incredible seeing Meta release a lightweight model for developers who want a fast, low-cost model they can rely on, and for teams cutting costs by running it as the subagent in their harness. Muse Ember is purpose-built for, and excels at, subagent work like coding and document analysis.”

Author

Role

Founder

Company

“Muse Ember showed an encouraging ability to tackle complex engineering tasks in our early testing. We see strong potential for it to become a valuable tool for engineering teams.”

Author

Role

SVP, Global Partnerships

Company

Oracle

“What stood out in our testing was speed. Muse Ember ran up to 30% faster at the tail than the comparable model we tested, and it handled broad, multi-repo investigation tasks well. It’s a strong fit for fast, high-throughput agentic workloads.”

Author

Role

CTO

Company

“Muse Ember completed every coding quest in our testing with the lowest latency of any model we tested: a p95 of 2.2 seconds, about 4x to 13x faster than leading open-weight models, and with 36–48% fewer output tokens. We’re excited about what Meta Superintelligence Labs is building next and about bringing it to our customers.”

Author

Role

Inference Lead

Company

Agents

Agents spend most of their tokens on many small, structured steps, and Muse Ember is built for exactly that. It can carry a business workflow end to end, pulling records, calling tools in sequence, and updating systems across apps.

With a 1M-token context window and up to 512K output tokens, it can take in a full codebase or document set in a single call.

AutomationBench-AA results at max reasoning, reproduced from the doc's current chart.. Score

AutomationBench-AAMuse Ember(max)GPT-6 Luna(max)Claude Haiku 5.5(max)0%10%20%30%40%Score37.8%20.2%13.9%

AutomationBench-AAMuse Ember(max)GPT-6 Luna(max)ClaudeHaiku 5.5(max)0%10%20%30%40%Score37.8%20.2%13.9%

Artificial Analysis | November 9, 2026. Strict pass@1.

Agents' Last Exam results at max reasoning, reproduced from the doc's current chart.. Score

Agents' Last ExamMuse Ember(max)GPT-6 Luna(max)Claude Haiku 5.5(max)0%10%20%30%40%50%60%70%Score61.6%57.1%57.0%

Agents' Last ExamMuse Ember(max)GPT-6 Luna(max)ClaudeHaiku 5.5(max)0%10%20%30%40%50%60%70%Score61.6%57.1%57.0%

Meta internal evaluation. Linux subset.

Personal Superintelligence Index results at max reasoning, reproduced from the doc's current chart.. Score

Personal Superintelligence IndexMuse Ember(max)GPT-6 Luna(max)Claude Haiku 5.5(max)0%10%20%30%40%50%Score46.8%37.0%36.1%

Personal Superintelligence IndexMuse Ember(max)GPT-6 Luna(max)ClaudeHaiku 5.5(max)0%10%20%30%40%50%Score46.8%37.0%36.1%

Internal composite index of personal and knowledge work.

Coding

Muse Ember is built for the coding subtasks an agent sends off hundreds of times a session: reading files, running tests, making targeted fixes.