← Model list

Muse Glimmer 30B

meta·meta/muse-glimmer-30b·GA·Open weights·muse series·NEW

Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.

At a glance
Reference price$0.35 / $1.50 per 1M
Cache read $0.04 · Fireworks AI
Lowest paid$0.20 / $0.80
EmpirioLabs AI Gateway · 1.7× spread
2 more $0 channels
Context131,072
Output limit131,072
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF
Knowledge cutoff2026-01-04
Released / updated2026-08-10 / 2026-08-10

Quality & performance

Artificial Analysis doesn't cover this model (267 of 2059 have data). Quality data comes from independent evals covering widely used models.

Available at 12 providers10 with public prices · 2 free

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
NvidiaFirst-partyFree131,072131,072
RequestyGatewayFree131,07220,480
EmpirioLabs AIGateway$0.20$0.80$0.05131,07232,768
OpenRouterGateway$0.30$1.10$0.04131,072131,072
Kilo GatewayGateway$0.30$1.10$0.04131,072131,072
Eden AI
deepinfra/meta-models/Muse-Glimmer-30B
Gateway$0.30$1.20$0.04131,072131,072
Eden AI
flexai/Muse-Glimmer-30B
Gateway$0.30$1.20131,072131,072
NanoGPTGateway$0.35$1.50$0.04131,072131,072
Vercel AI GatewayCloud$0.35$1.50$0.04131,072131,072
Eden AI
together_ai/meta-models/Muse-Glimmer-30B
Gateway$0.35$1.50$0.04131,072131,072
Eden AIGateway$0.35$1.50$0.04131,072131,072
Fireworks AICloud$0.35$1.50$0.04131,072131,072

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.

Your usage cost

1EmpirioLabs AI$62.00
2OpenRouter$83.80
3Kilo Gateway$83.80
4Eden AI$88.80
5NanoGPT$107.80
6Vercel AI Gateway$107.80

2 more channels offer $0 (Nvidia, Requesty); free tiers usually have rate limits and no SLA, excluded from ranking.

The cheapest paid channel is the only channel.

Benchmark9 items

NameConditionsScoreMetricSource
MCP Atlasvariant: public75.5success rateSource ↗
DeepSearch QA74.6Source ↗
SWE-Bench Pro51.2resolve rateSource ↗
SWE-Bench Verified76resolve rateSource ↗
Terminal-Benchvariant: with terminus2 · v2.151.7success rateSource ↗
OSWorld-Verified65.9success rateSource ↗
AIME 202694.7accuracySource ↗
GPQA Diamondvariant: AA83.5accuracySource ↗
CharXiv Reasoning78.8accuracySource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Reasoning control

effort = noneeffort = minimaleffort = loweffort = mediumeffort = higheffort = maxbudget_tokenseffort = xhigh

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$1.50$02026-08-112026-08-172026-08-11 · Input list $0.352026-08-12 · Input list $0.352026-08-13 · Input list $0.352026-08-17 · Input list $0.352026-08-11 · Output list $1.502026-08-12 · Output list $1.502026-08-13 · Output list $1.502026-08-17 · Output list $1.502026-08-11 · Min blended $0.6372026-08-12 · Min blended $0.5252026-08-13 · Min blended $0.5252026-08-17 · Min blended $0.35