MetaMuse Ember
No cheaper ranked model scores higher: it is on the price frontier.
As of 9 Nov 2026, Muse Ember by Meta is not yet ranked on the BenchLeader Index, which needs at least five independent results across two categories. API pricing: $0.100 per 1M input tokens and $0.500 per 1M output tokens ($0.200 blended at 3:1), list price on 9 Nov 2026. It streams about 190 tokens/s (OpenRouter). Released 9 Nov 2026.
Key facts (as of 9 Nov 2026)
- Muse Ember API pricing: $0.100 per 1M input tokens and $0.500 per 1M output tokens ($0.200 blended at 3:1) (list price, 9 Nov 2026).
- Muse Ember streams about 190 tokens/s (OpenRouter, 9 Nov 2026).
- Muse Ember was released on 9 Nov 2026 (Meta).
In detail. Muse Ember is a Meta proprietary reasoning model, released 9 Nov 2026. It is not yet ranked: the index needs at least five independent results across two categories. At $0.200 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 189 tokens per second puts it in the fastest quarter, with a first token in 1.8 s. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 9 Nov 2026.
- Blended price
- $0.200/M
- $0.100 in · $0.500 out
- via Artificial Analysis
- Output speed
- 189 tok/s
- OpenRouter traffic, 7-day median; not yet measured by Artificial Analysis
- First answer
- 1.81 s
- Context
- 1.0M
- via OpenRouter
- Full answer
- –
- Cost per task
- $0.116
- average task, AA Index
- Tokens per task
- 72k
- 61% of it reasoning
- Released
- 9 Nov 2026
- Reasoning51
- Coding51
- Knowledge49
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| xhighbest | – | – | 189 tok/s | 1.81 s | $0.0002 | Coding 51 · Knowledge 49 · Reasoning 51 |
| max | – | – | 189 tok/s | 1.81 s | $0.0002 | Coding 57 · Knowledge 49 · Reasoning 52 |
| not stated | – | – | 189 tok/s | 1.81 s | $0.0002 |
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | xhigh | max | Source |
|---|---|---|---|
| HLE-Diamond (AA) | 23.6%#65 | 25.1%#57 | Artificial Analysis |
| CritPt v2 hard subset (AA) | 38.5%#57 | 39.0%#55 | Artificial Analysis |
Coding
| Benchmark | xhigh | max | not stated | Source |
|---|---|---|---|---|
| Vibe Code Bench v1.1not in index | – | 83.7%#23 | – | Vals AI |
| Terminal-Bench 4.0 (AA)not in index | 35.4%#48 | 40.9%#42 | – | Artificial Analysis |
| Design Arena Websitenot in indexeffort not stated | – | – | 1330#8 | Design Arena |
| FrontierCode 2.0 (AA) | 22.0%#57 | 27.2%#42 | – | Artificial Analysis |
Agents & tools
| Benchmark | xhigh | max | Source |
|---|---|---|---|
| Finance Agent v2.1not in index | – | 63.8%#15 | Vals AI |
| Harvey's Legal Agent Benchmark v1.1not in index | – | 25.0%#2 | Vals AI |
| Legal Research Benchnot in index | – | 49.0%#10 | Vals AI |
| Tax Agent Benchnot in index | – | 38.2%#14 | Vals AI |
| Terminal-Bench 4.0 (Vals)not in index | – | 39.4%#16 | Vals AI |
| AutomationBenchnot in index | 66.7%#23 | 68.1%#17 | Artificial Analysis |
| GDP.pdf (AA)not in index | 22.4%#65 | 22.8%#59 | Artificial Analysis |
| AA-Briefcase v1.1not in index | 1527#34 | 1521#38 | Artificial Analysis |
| Terminal-Bench-Science 0.1 (AA)not in index | 17.6%#47 | 20.0%#44 | Artificial Analysis |
Knowledge
| Benchmark | xhigh | max | Source |
|---|---|---|---|
| AA-Omniscience v1.1 | 7.7#62 | 7.4#63 | Artificial Analysis |
| Excel Modeling Benchmark v1.1not in index | – | 72.2%#12 | Vals AI |
| AA-Omniscience v1.1: accuracynot in index | 21.4%#107 | 21.1%#112 | Artificial Analysis |
| AA-Omniscience v1.1: non-hallucinationnot in index | 82.7%#5 | 82.6%#6 | Artificial Analysis |
Composite
| Benchmark | xhigh | max | Source |
|---|---|---|---|
| AA Intelligence Index v5.0not in index | 34.1#50 | 36.0#43 | Artificial Analysis |
| Vals Indexnot in index | – | 60.5#13 | Vals AI |
| Vals Finance Indexnot in index | – | 60.9%#11 | Vals AI |
API pricing: where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Meta | 189 tok/s | 1.81 s | $0.100 | $0.500 | 1.0M | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | – | 3.4 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0015 | – | 5.0 s |
| Code edit | 6,000 / 1,500 | $0.0014 | – | 9.7 s |
| Agentic coding session | 60,000 / 4,000 | $0.0080 | – | 23.0 s |
| Structured extraction | 2,000 / 200 | $0.0003 | – | 2.9 s |
See also
Questions people ask
- Why does Muse Ember have no BenchLeader Index score?
- The index needs at least five independent results across at least two categories, and as of 9 Nov 2026 Muse Ember does not have them yet. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
- How much does Muse Ember cost per 1M tokens?
- As of 9 Nov 2026, Muse Ember costs $0.100 per 1M input tokens and $0.500 per 1M output tokens, $0.200 blended (three input tokens per output token).
- When was Muse Ember released?
- Meta released Muse Ember on 9 Nov 2026. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 9 Nov 2026. Compare these configurations.
Cite as: BenchLeader, “Muse Ember: benchmarks, pricing, speed and rank”, data as of 9 Nov 2026, https://www.benchleader.com/models/muse-ember.