i
News
News · 2026-11-09

Meta releases Muse Ember, a small model for everyday AI-agent tasks

Dataist Dataist

Meta has introduced Muse Ember, its smallest and most cost-efficient model. It is designed for routine AI-agent work: calling tools, handling small coding tasks, routing requests, extracting data and processing long documents. Meta says Muse Ember delivers more capability per dollar than Claude Haiku 5.5 and GPT-6 Luna at every reasoning level. In an Artificial Analysis test, a task costs about $0.14: the larger Muse Spark 1.3 scored nearly as much — 37 versus 36 — but costs $2.39. In Meta’s own tests, Muse Ember leads in most cases. But on real-world coding tasks, Haiku outperformed it on the FrontierCode 2.0-AA benchmark: 34.9% versus 27.2%.

Cover: Meta releases Muse Ember, a small model for everyday AI-agent tasks

Price and benchmark results

Muse Ember scored 36 on the Artificial Analysis Intelligence Index, at a cost of about $0.14 per task. Muse Spark 1.3 scored 37, but a task with it costs $2.39.

Meta’s Muse Ember achieves a score of 36 on the Artificial Analysis Intelligence Index at a particularly low cost per task. | Image: Meta via Artificial Analysis

Meta’s Muse Ember achieves a score of 36 on the Artificial Analysis Intelligence Index at a particularly low cost per task. | Image: Meta via Artificial Analysis

Source: the-decoder.com

Meta says Muse Ember delivers more capability per dollar than Claude Haiku 5.5 and GPT-6 Luna at every reasoning level. Results vary by task type: Muse Ember leads in most of Meta’s own tests, while Haiku scored 34.9% versus Muse Ember’s 27.2% on the FrontierCode 2.0-AA benchmark for real-world coding tasks.

Muse Ember36 points for $0.14
Muse Spark 1.337 points for $2.39

Access and capabilities

The model works with text, images, audio and video, and can process up to one million tokens at a time. It is already available through Meta Model API and Muse Code. Pricing is $0.10 per million input tokens and $0.50 per million output tokens.

Dataist’s take: the results show that Muse Ember’s low cost does not guarantee it will lead across every task type: it fell behind Haiku on a real-world coding test. Models like these should therefore be compared with the specific work they are intended to do in mind.

Daily AI news

Every day we pick what actually matters in AI and explain it plainly — no hype, no filler. Subscribe if you want to follow where the industry is going.

Only what matters — every day

Follow on X