MetaMuse Glimmer
Muse Glimmer is a Meta open-weights reasoning model, released 10 Aug 2026. Its best configuration ranks #248 of 785 on the BenchLeader Index at 54.7 ±7.8, in the upper half. The ± is the point: 390 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in maths (62) and lowest in coding (46). At $0.525 per million tokens blended it is mid-priced. Output speed of 113 tokens per second puts it faster than most, with a first token in 0.8 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 23 Oct 2026.
- Blended price
- $0.525/M
- $0.300 in · $1.20 out
- Output speed
- 113 tok/s
- OpenRouter traffic, 7-day median; not yet measured by Artificial Analysis
- First answer
- 0.79 s
- Context
- 131k
- Full answer
- –
- Cost per run
- –
- Released
- 10 Aug 2026
- Overall index55
- Reasoning61
- Coding46
- Maths62
- Human preference60
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 |
|---|---|---|---|---|---|---|
| high | 51.4 | #347 | 119 tok/s | 18 s | $0.0005 | Agents & tools 52 · Coding 47 · Composite 49 · Knowledge 47 · Long context 67 · Multimodal 57 · Reasoning 45 |
| not statedbest | 54.7 | #248 | 113 tok/s | 0.79 s | $0.0005 | Coding 46 · Human preference 60 · Maths 62 · Reasoning 61 |
How its index has moved
9 Sept 2026 to 21 Oct 2026- not stated
- high
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
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 | high | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Hard Promptseffort not stated | – | 1447#113 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 83.5%#156 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 22.0%#203 | – | Artificial Analysis | |
| CritPt | 2.6%#189 | – | Artificial Analysis |
Coding
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| SciCode | 44.9%#142 | – | SciCode | |
| Arena Codingeffort not stated | – | 1478#108 | Arena (LMArena) | |
| Arena WebDeveffort not stated | – | 1355#109 | Arena (LMArena) | |
| SciCode (AA)not in index | 44.9%#148 | – | Artificial Analysis | |
| Terminal-Bench 4.0 (AA)not in index | 0.5%#139 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 51.7%#115 | – | Artificial Analysis | |
| FrontierCode 1.1 (AA)not in index | 13.9%#63 | – | Artificial Analysis |
Agents & tools
| Benchmark | high | Source | Trend |
|---|---|---|---|
| GDPval-AA v2.1not in index | 14.2%#107 | Artificial Analysis | |
| τ³-Banking (AA) | 23.5%#86 | Artificial Analysis | |
| Analyst Agent (AA)not in index | 20.0%#35 | Artificial Analysis | |
| AutomationBenchnot in index | 6.8%#158 | Artificial Analysis | |
| GDP.pdf (AA)not in index | 10.0%#133 | Artificial Analysis | |
| EnterpriseOps-Gymnot in index | 34.7%#34 | Artificial Analysis | |
| Terminal-Bench-Science 0.1 (AA)not in index | 0.0%#72 | Artificial Analysis |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Maths | 1446#82 | Arena (LMArena) |
Knowledge
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -32.9#290 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 27.0%#209 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 18.1%#303 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Instruction Followingnot in index | 1414#118 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1425#118 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1365#153 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1432#111 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1422#130 | Arena (LMArena) |
Multimodal
| Benchmark | high | Source | Trend |
|---|---|---|---|
| MMMU-Pro | 74.3%#112 | Artificial Analysis |
Long context
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA-LCR | 83.3%#21 | Artificial Analysis |
Composite
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 17.5#262 | Artificial Analysis |
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 |
|---|---|---|---|---|---|---|
| DeepInfra | 100 tok/s | 0.87 s | $0.300 | $1.20 | 131k | bf16 |
| Together | 96 tok/s | 0.34 s | $0.350 | $1.50 | 131k | – |
Price history
The price above, per 1M tokens, every day it changed.
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 | $0.0005 | – | 3.4 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0043 | – | 6.1 s |
| Code edit | 6,000 / 1,500 | $0.0036 | – | 14.1 s |
| Agentic coding session | 60,000 / 4,000 | $0.023 | – | 36.2 s |
| Structured extraction | 2,000 / 200 | $0.0008 | – | 2.6 s |
More from Meta
See also
Questions people ask
- What are Muse Glimmer's benchmark scores?
- As of 25 Oct 2026, Muse Glimmer scores 54.7 ±7.8 on the BenchLeader Index, strongest in maths (62) and weakest in coding (46). That rests on 5 independent benchmark results, and every score on this page links to its publisher.
- How does Muse Glimmer rank?
- Muse Glimmer is #248 of 785 ranked configurations on 25 Oct 2026. 390 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Muse Glimmer cost per 1M tokens?
- As of 25 Oct 2026, Muse Glimmer costs $0.300 per 1M input tokens and $1.20 per 1M output tokens, $0.525 blended (three input tokens per output token). The list price last changed on 6 Oct 2026.
- When was Muse Glimmer released?
- Meta released Muse Glimmer on 10 Aug 2026. BenchLeader has results for it at 2 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Muse Glimmer: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/muse-glimmer.