modelbenchmark.io

Muse Glimmer 30B

Meta · muse-glimmer-30b

Compare

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.

Specification

most-agreed values

Context
131K
Max output
118K
Released
2026-08-10
Knowledge cutoff
2026-01-04
Retires
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.35
Output
$1.50
Cache read
$0.04

Available from 14 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nvidiafreefree131K131K
requestyfreefree131K20K
empiriolabs$0.20$0.80$0.05131K33K
deepinfra$0.30$1.20$0.04131K
edenai
$0.30
$0.35
$1.20
$1.50
$0.04131K131K
edenai · flexai$0.30$1.10131K131K
kilo$0.30$1.10$0.04131K118K
fireworks_ai$0.35$1.50$0.04131K16K
fireworks-ai · accounts$0.35$1.50$0.04131K131K
nano-gpt$0.35$1.50$0.04131K118K
nano-gpt · tee$0.35$1.50$0.04131K131K
openrouter$0.35$1.50$0.04131K118K
together_ai$0.35$1.50$0.04131K
vercel$0.35$1.50$0.04131K131K

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02openrouterInput$0.30
2026-09-13openrouterInput$0.35
2026-09-02openrouterOutput$1.20
2026-09-03openrouterOutput$1.10
2026-09-13openrouterOutput$1.50

More from Meta

most-hosted first