Meta · 2026-11-09 · major
Muse Ember — Meta's smallest Muse model for agent sub-tasks
In short
Muse Ember is Meta's smallest and most efficient Muse model, built for tool calls, coding subtasks, routing and extraction. It costs $0.10 per million input tokens and $0.50 per million output tokens, with a 1M-token context.
Meta · 2026-11-09 · 3 sources checked
Muse Ember is Meta's cheapest Muse model, made to run the many small steps inside an AI agent.
Quick facts
| Maker | Meta Superintelligence Labs |
|---|---|
| Released | 9 November 2026 |
| API model ID | muse-ember-1.0 |
| Context window | 1M tokens · up to 512K output |
| Price | $0.10 input / $0.50 output per 1M tokens |
| Availability | Meta Model API and Muse Code |
| Weights | Not released (proprietary) |
Benchmarks
Pricing
| Input | $0.10 / 1M tokens |
|---|---|
| Cached input | $0.004 / 1M tokens |
| Output | $0.50 / 1M tokens |
What is it?
Muse Ember is the smallest and most efficient model in Meta's Muse family, released on 9 November 2026. It targets agent work such as tool calls, coding subtasks, routing, extraction and long-document parsing, and it reads up to 1M tokens and writes up to 512K.
How does it work?
Meta says Muse Ember is the first model it trained using Muse Wonder Full, its larger frontier setting. Developers call it through the Meta Model API as muse-ember-1.0, including the Responses API, or use it inside Muse Code, at several reasoning settings.
Why does it matter?
Agents make many small model calls, so price per step decides what a long agent run costs. At $0.10 in and $0.50 out per million tokens, Meta pitches Ember against Claude Haiku 5.5 and GPT-6 Luna, and reports it ahead on most of its agent tests, including 33.8% on OSWorld 2.1 (binary).
Who is it for?
developers building high-volume AI agents
Frequently asked questions
- How much does Muse Ember cost on the Meta Model API?
- Muse Ember costs $0.10 per million input tokens, $0.004 per million cached input tokens and $0.50 per million output tokens on the Meta Model API. Meta cites Artificial Analysis figures of about $0.14 per Intelligence Index task for Muse Ember, against $2.39 per task for Muse Spark 1.3.
- How does Muse Ember compare to Claude Haiku 5.5 and GPT-6 Luna?
- Meta's own table at max reasoning puts Muse Ember ahead of Claude Haiku 5.5 and GPT-6 Luna on most agent tests, for example 37.8% on AutomationBench-AA against 13.9% and 20.2%. Claude Haiku 5.5 leads on FrontierCode 2.0-AA, 34.9% against Muse Ember's 27.2%, and the two tie at 36 on the Artificial Analysis Index.
- Are Muse Ember's weights open?
- No. Meta's announcement makes Muse Ember available through the Meta Model API as muse-ember-1.0 and inside Muse Code, and it does not mention open weights or a parameter count. Meta's open-weight Muse models are the separate Muse Spark Open and Muse Glimmer lines.
- What is Muse Ember meant to be used for?
- Meta positions Muse Ember for the high-volume steps of agent workflows: tool calls, coding subtasks, routing, extraction and parsing long documents. For the hardest problems Meta suggests pairing it with Muse Wonder: Muse Wonder plans the work and Muse Ember handles the volume of cheap steps.
Try it
Model ID muse-ember-1.0 on the Meta Model API (developer.meta.com/ai)