Meta releases Muse Ember, a low-cost model for high-volume agent work

What happened

Muse Ember costs $0.10 per million input tokens and $0.50 per million output tokens.

It has a 1M-token context window and accepts text, images, audio and video.

It is available through the Meta Model API (as muse-ember-1.0), Muse Code, Google Cloud and Oracle Cloud.

Meta pitches it for tool calls, coding subtasks, routing and extraction, running alongside a larger model rather than replacing it.

As with earlier Muse releases, the weights remain private, so the model is used through Meta's API and cloud partners rather than run locally.

Why it matters for creators

Cheap models that read video and audio make bulk tasks such as tagging, captioning and clip selection less costly.

These jobs often involve large volumes of footage and audio. A small, low-cost model that accepts media directly can handle them, leaving larger and more expensive models for the steps that need them, such as writing or planning.

The 1M-token context window also gives room for long inputs in a single request.