VerdictPal
Join

Models / Meta

Muse Glimmer

Muse · Released 2026-08-10

Meta's 30B open-weight local-agent model (10 Aug 2026). Artificial Analysis Intelligence Index 35.1 on the high-effort endpoint. AA hosted list price $0.325 / $1.35 per 1M tokens; weights are Apache 2.0.

On-device Muse variant for always-on local agents, function calling, and local coding. Prefer sourced ledger rows over launch claims; AA scores sync from the public snapshot.

#94 of 166 on AA Index · current snapshot
35 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite48
ContextUnlisted
Input / 1M$0.33
Output / 1M$1.4
Knowledge cutoffunpublished
Output speed101 t/s
TTFT (default API)0.46s
Statussolid

Family profile

Best published score in each covered benchmark family.

65/100 avg
Reasoning
84
Coding
44
Agentic
52
Long context
80

4 tested benchmark families

Editor's note
Recorded from the 2026-08-13 AA snapshot (slug muse-glimmer, high effort, index 35.1) plus the Meta Superintelligence Labs launch post. Editorial depth still needs a desk pass — scores below are ledger-backed, not hands-on claims. This is the local 30B Muse lane, not a Spark replacement.
Benchmark placements

Where Muse Glimmer places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
AA-LCRLong context#7/ 17980Artificial Analysis2026-09-01
AA output speedPerformance#38/ 185Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#48/ 10624Artificial Analysis2026-09-01
AA time to first tokenPerformance#79/ 185Artificial Analysis2026-09-01
Terminal-BenchAgentic#87/ 18452Artificial Analysis2026-09-01
SciCodeCoding#90/ 18044Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#108/ 18435Artificial Analysis2026-09-01
GPQA DiamondReasoning#115/ 18584Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#115/ 18022Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where Muse Glimmer shows up, it's highlighted.

ReasoningMuse Glimmer · 84
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingMuse Glimmer · 44
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticMuse Glimmer · 52
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextMuse Glimmer · 80
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for Muse Glimmer in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context80%Artificial Analysis2
AA output speedPerformance101 t/sArtificial Analysis2
AA time to first tokenPerformance0.46sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning35.1%Artificial Analysis2
GPQA DiamondReasoning83.5%Artificial Analysis2
Humanity's Last ExamReasoning22%Artificial Analysis2
SciCodeCoding43.6%Artificial Analysis2
τ³-Bench BankingAgentic23.51%Artificial Analysis2
Terminal-BenchAgentic51.69%Artificial Analysis2
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning84
Coding44
Agentic52
Long context80
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Muse Glimmer. Scores link to the original source; gaps mean no public row exists yet.

9 sourced rows

9 of 37 catalog benchmarks have a sourced row for Muse Glimmer.

Reasoning

Coding

Agentic

Long context

Performance

Changelog
  1. Muse Glimmer open-weighted (Apache 2.0, 30B). AA high-effort index 35.1; hosted list $0.325 / $1.35 per 1M tokens.
Sources