# Meta releases Muse Ember model

The multimodal model offers a 1M-token context window and costs $0.10 per million input and $0.50 per million output tokens on the Meta Model API.

- Supports native inputs across text, images, audio, and video.
- Built for agent workflows including sequential tool calling, test execution, and targeted code fixes.
- Matches Claude Haiku 5.5 on the Artificial Analysis Intelligence Index at about one-third the task cost and outscores GPT-6 Luna across reasoning settings.

## Why it matters

Developers building agentic workflows have a lower-cost multimodal option with a large context window.

## Sources

- [Meta AI: Meta releases Muse Ember model](https://x.com/AIatMeta/status/2119853591703998787)

---

Summarized by dstilled on 2026-11-09. https://dstilled.ai/story/70d834e9-cdd8-499d-ac5a-5bf08479b391
