Muse Ember is Meta's smallest and most efficient reasoning model, built for the work agents repeat all day: tool calls, coding subtasks, routing, extraction, and long-document parsing. It pairs with Muse Wonder or Muse Spark, which plan while Muse Ember handles the volume.Muse Ember by meta is offered by 10 providers on this page. Its official list price is $0.10 per 1M input tokens and $0.50 per 1M output tokens. No paid channel is priced below the official list price. Prices vary about 1.7× across the providers that host it.
Artificial Analysis rates it 36 on the Intelligence Index, with 48.7 for Strategy & Ops, its strongest domain. Against its lowest blended price of $0.20 per 1M, that is roughly 180 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 604 tokens per second, with 8.63s to the first token. Running one task of the Artificial Analysis suite costs about $0.1359, which reflects how many tokens its reasoning consumes rather than the unit price alone.
The context window is 1,048,576 tokens, with an output limit of 524,288 tokens. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, Video, PDF, and Audio.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| MetaOfficial | First-party | $0.10 tiered >200K: $0.30 | $0.50 | $0.004 | — | 1,048,576 | 524,288 | |
| DevPass (LLM Gateway) | Gateway | $0.10 | $0.50 | $0.004 | — | 1,048,576 | 524,288 | |
| LLM Gateway | Gateway | $0.10 | $0.50 | $0.004 | — | 1,048,576 | 524,288 | |
| OpenCode | Gateway | $0.10 tiered >200K: $0.30 | $0.50 | $0.004 | $0 | 1,048,576 | 270,000 | |
| Cloudflare AI Gateway | Cloud | $0.10 | $0.50 | $0.004 | — | 1,048,576 | 524,288 | |
| OpenRouter meta/muse-ember | Gateway | $0.10 | $0.50 | $0.004 | $0.13 | 1,048,576 | 943,718 | |
| Vercel AI Gateway meta/muse-ember | Cloud | $0.10 | $0.50 | $0.004 | — | 1,048,576 | 1,048,576 | |
| NanoGPT meta/muse-ember | Gateway | $0.10 | $0.50 | $0.004 | — | 1,048,576 | 1,048,576 | |
| Kilo Gateway meta/muse-ember | Gateway | $0.10 | $0.50 | $0.004 | — | 1,048,576 | 943,718 | |
| Kenari muse-ember-1-0 | Gateway | $0.17 | $0.86 | $0.0086 | — | 1,048,576 | 524,288 |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Price history accumulates from each data sync; currently only 1 sample(s) (2026-11-10). Each future sync adds a point, and once accumulated a line is drawn here.
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| AutomationBench | variant: max effort; Artificial Analysis AutomationBench-AA | 37.8 | strict pass@1 | Source ↗ |
| Agents' Last Exam | harness: Meta internal evaluation framework · variant: max effort · dataset: Linux subset | 61.6 | score | Source ↗ |
| Terminal-Bench | harness: mini-swe-agent · variant: max effort · v4.0 | 40.9 | accuracy | Source ↗ |
| FrontierCode | variant: max effort; Artificial Analysis FrontierCode-AA · v2.0 | 27.2 | score | Source ↗ |
| GDP.pdf | variant: max effort; Artificial Analysis GDP.pdf-AA | 22.8 | score | Source ↗ |
| OSWorld | variant: max effort · v2.1 | 69.5 | mean partial score | Source ↗ |
| OSWorld | variant: max effort · v2.1 | 33.8 | binary completion rate | Source ↗ |
| PerceptionBench | harness: Meta internal evaluation framework · variant: max effort; with tools | 80.6 | score | Source ↗ |
| Artificial Analysis Intelligence Index | variant: max effort · v5 | 36 | index score | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.
Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and scale (percent correct, points or an index), so rows are not comparable with one another.