Muse Ember is a large language model made by Meta, added to the catalog on 9 Nov 2026. It has a context window of 1,048,576 tokens and costs $0.10 per 1M input tokens and $0.50 per 1M output tokens. It accepts text, images, files, audio and video and produces text. Typical uses: reasoning over multimodal inputs, coding agent subtasks and tool-assisted workflows.
Price checked · source: OpenRouter
What Muse Ember is
Meta: Muse Ember is a Meta model designed for complex reasoning and recurring agent work. Its catalog description names tool calls, coding agent subtasks, and routing as examples of that work. The model accepts text, images, video, files, and audio, and produces text. It supports reasoning, tools, structured outputs, and response formats.
Its context length is 1,048,576 tokens, and its maximum output is 943,718 tokens. These limits make the model suitable for tasks that use substantial context or call for extended text output, subject to the limits stated in the catalog. The context limit describes how much material can be held, while the output limit describes how much text it can produce. Its listed input formats allow a workflow to provide material as text, images, video, files, or audio; the listed output is text.
The combination of reasoning and tool support fits workflows involving tool calls, while structured outputs and response formats are relevant where a task expects text in a defined form. The catalog also specifically identifies coding agent subtasks and routing among the model’s intended work. Its multimodal input coverage allows workflows to bring different listed kinds of material into a task, while generated responses remain text. These stated areas of focus make Muse Ember relevant to recurring reasoning tasks across its supported input types.
Key features
- Accepts text, image, video, file, and audio inputs
- Produces text output
- Supports a context length of 1,048,576 tokens
- Supports tools, structured outputs, and response formats
- Supports reasoning
Typical uses
- Reasoning over multimodal inputs
- Coding agent subtasks
- Tool-assisted workflows
- Routing tasks
- Generating structured text responses
Muse Ember pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.10 |
| Output, per 1M tokens | $0.50 |
| Cached input read, per 1M tokens | $0.004 |
| Cache write, per 1M tokens | $0.125 |
| Web search, per search | $0.0025 |
No price change since it was added on 9 Nov 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0002 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0029 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0040 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.016 |
Muse Ember specifications
| Model id | meta/muse-ember |
|---|---|
| Version | muse-ember-20261109 |
| Context window | 1,048,576 tokensAbout 786,432 English words |
| Max output | 943,718 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 9 Nov 2026 |
| Input types | text, image, video, file, audio |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | Yes |
| Content moderation | Applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Muse Ember yet.
Compare Muse Ember
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.