Inference
EN
Meta

Muse Ember

Muse Ember is a reasoning model from Meta. 9 benchmarks count toward its index, in 4 categories.

  • New
  • Closed weights
  • Reasoning
  • Input: text · image · video · file · audio
  • Released 2026-11
RankedEnough results for an index.
IndexAll categories, out of 100.64.8±8.4Median 51.0
CoverageIndex weight that has results.70%
SpeedOutput tokens per second.158/sMedian 67/s
First tokenSeconds to the first output token.2.1 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.10Median $0.60
Output / 1MUS dollars per 1M output tokens.$0.50Median $2.55
ContextMaximum tokens in one request.1.05MMedian 262K
EloLMArena rating and rank.—Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
meta/muse-ember
Max outputMaximum output tokens per request.
944K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
<$0.01
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.6 counted71.421 / 218
CodingCode writing and repair.1 counted72.5—
ReasoningLogic problems and puzzles.1 counted63.4—
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.1 counted37.3—
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.None counted——

Results

9 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Vibe Code Bench v1.1Coding83.7% ±2.821 / 123
Agent Arena steerabilityAgentic12.010 / 54
Terminal-Bench 4.0Agentic39.4% ±2.315 / 61
Artificial Analysis AutomationBenchAgentic68.1%8 / 23
Artificial Analysis Terminal-Bench v4.0Agentic40.9%13 / 22
Artificial Analysis GDP.pdfAgentic22.8%13 / 22
Medical Long Context Reasoning (MLCR-AA)Reasoning18.3%16 / 22
Artificial Analysis Omniscience AccuracyKnowledge21.1%126 / 164
Agent Arena task outcomeAgentic−7.642 / 54
OSWorld 2.1 (all 108 tasks)Agentic69.5%10 / 11
HLAB-Vals-1.1Legal25.0% ±3.72 / 35
Legal Research BenchLegal49.0% ±3.510 / 90
Tax Agent BenchFinance38.2% ±3.513 / 82
PerceptionBench with toolsMultimodal80.6%—
Agents' Last Exam, Linux subsetAgentic61.6%—
Artificial Analysis AutomationBench, strict pass@1Agentic37.8%—
FrontierCode 2.0, Artificial Analysis evaluationCoding27.2%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseVals AINo licence statedLMArenaCC BY 4.0

History

5 releases
Index64.4 → 64.8
11 Nov 202612 Nov 2026
Output / 1M$0.50 → $0.50
11 Nov 202612 Nov 2026

Questions

How much does Muse Ember cost?
$0.10 per 1M input tokens, $0.50 per 1M output tokens.
How good is Muse Ember?
Its index is 64.8 out of 100. That is place 46 of 283 ranked models.
What is Muse Ember best at?
Agentic: 71.4 out of 100, place 21 of 218.
How fast is Muse Ember?
158 output tokens per second, 2.1 s to the first token.
What is the context window of Muse Ember?
1.05M tokens.
Is Muse Ember open source?
No. Its weights are not published.
When was Muse Ember released?
November 2026.

Badge

Muse Ember on Inference360: place 46 of 283, index 64.8
HTML
<a href="https://inference360.ai/models/muse-ember"><img src="https://inference360.ai/badges/muse-ember.svg" alt="Muse Ember on Inference360: place 46 of 283, index 64.8" height="20"></a>
Markdown
[![Muse Ember on Inference360: place 46 of 283, index 64.8](https://inference360.ai/badges/muse-ember.svg)](https://inference360.ai/models/muse-ember)

Compare with

Muse Spark 1.264.9Muse Spark 1.164.6Muse Spark Open65.1

Lower price

dots3-note Preview63.4 · FreeGLM-5.3-Flash63.1 · $0.078/1MLing 3.1 Flash67.2 · $0.135/1M

More from Meta

All models
Muse Spark Open65.1Muse Wonder Full77.3Muse Wonder73.1Muse Spark 1.3 ContributorUnrankedMuse Spark 1.370.1