Muse Glimmer-30B
Muse Glimmer-30B is Meta Superintelligence Labs' open-weight, dense multimodal model for autonomous agentic work on consumer hardware. Its 29.6-billion-parameter causal transformer includes a roughly 1.8B-parameter ViT-G/14 perception encoder, accepts interleaved text and images, produces text, and supports a 131,072-token context window. It combines controllable multi-step reasoning, schema-based tool use, coding, failure recovery, and compatibility with agent scaffolds such as OpenClaw and Hermes Agent. Meta also releases 4-bit variants sized for 24 GB and 32 GB systems and a DFlash speculative-decoding drafter. The model is trained on more than 100 languages and released under Apache 2.0.
Identité
| Organisation | Meta |
|---|---|
| Pays | — |
| Famille | — |
| Type | llm |
| Date de sortie | 2026-08-10 |
| Licence | apache 2.0 |
| Licence commerciale | oui |
| Licence open source | oui |
| Poids ouverts | oui |
| Auto-hébergement (DGX Spark) | oui — ≤ 107 Md (Q8/FP8) |
| Modalités | image|text |
| Fenêtre de contexte | 131 072 |
| KV cache / token (Q8) | 26.6 ko/token |
| Paramètres totaux | 29.6 B |
| Paramètres actifs (session) | 29.8 B |
| Source des paramètres actifs | actifs = total (modèle dense, config HF (meta-models/Muse-Glimmer-30B)) ; total : config Hugging Face (meta-models/Muse-Glimmer-30B) |
| Curation des paramètres actifs | 2026-11-04 |
| Prix par session | 0.54 $ (1 000 000 tokens entrée + 200 000 sortie, prix le moins cher par fournisseur) |
| Tokens / tâche (Data Players) | 1.2 M |
| Tokens / tâche (Terminal-Bench) | 83.3 M (est.) |
| Ratio Terminal-Bench / Data Players | 69.4× |
| Coût / tâche (Terminal-Bench) | 4.54 $ (est.) |
| Tokens d'entraînement | — |
| Coupure de connaissance | 2026-01-04 |
| Créé le | 2026-08-11T05:14:33.941236Z |
| Mis à jour le | 2026-11-04T06:04:51.693146Z |
| Source | LLM Stats |
| URL | https://llm-stats.com/models/muse-glimmer-30b |
Prix par fournisseur
Statut du modèle (offres + auto-hébergement) : partiel — auto-hébergement DGX Spark (≤ 107 Md).
Toutes sources confondues : catalogue llm-stats, routeurs EUrouter/Cortecs et catalogues directs (OVHcloud, Scaleway, Infomaniak). Les prix sont convertis en USD pour la comparaison ; le prix publié et le taux BCE utilisé sont dans l'infobulle. Les prix inconnus restent indiqués « — ».
| Fournisseur | Source | Souveraineté | Entrée (/M tok) | Sortie (/M tok) | Quantification | Contexte | Latence (s) | Débit (tok/s) | Statut |
|---|---|---|---|---|---|---|---|---|---|
| DeepInfra | llm-stats | non | 0.30 $ | 1.20 $ | — | — | — | — | active |
Souveraineté qualifiée par offre (modèle × fournisseur) ; survolez le badge pour les critères.
SouverainetéScores par catégorie (top_scores)
| Catégorie | Score |
|---|---|
| chat | 0.80 |
| instruction_following | 0.80 |
| language | 0.80 |
| legal | 476.90 |
| long_context | 0.80 |
| math | 0.60 |
| multimodal | 0.70 |
| physics | 0.60 |
| reasoning | 48.30 |
| search | 0.70 |
| spatial_reasoning | 0.70 |
| structured_output | 0.80 |
| finance | 476.90 |
| frontend_development | 0.80 |
| general | 37.30 |
| grounding | 0.80 |
| healthcare | 0.80 |
| agents | 95.80 |
| biology | 0.60 |
| chemistry | 0.60 |
| code | 0.60 |
| coding | 0.50 |
| communication | 0.90 |
| tool_calling | 0.70 |
| vision | 0.70 |
Scores par benchmark
| Benchmark | Catégorie | Score | Max | Normalisé (API) | Vérifié | Auto-déclaré | Daté du |
|---|---|---|---|---|---|---|---|
| aa-lcr | reasoning | 0.80 | 1.00 | 0.80 | non | oui | 2026-11-04T06:04:51.883033Z |
| aime-2026 | math | 0.95 | 1.00 | 0.95 | non | oui | 2026-11-04T06:04:51.853111Z |
| arc-agi | reasoning | 0.67 | 1.00 | 0.67 | non | oui | 2026-11-04T06:04:51.933770Z |
| beam-128k | reasoning | 0.65 | 1.00 | 0.65 | non | oui | 2026-11-04T06:04:51.893033Z |
| charxiv-r | reasoning | 0.79 | 1.00 | 0.79 | non | oui | 2026-11-04T06:04:51.783046Z |
| ci-memories-coverage | agents | 0.65 | 1.00 | 0.65 | non | oui | 2026-11-04T06:04:51.821578Z |
| ci-memories-violation | agents | 0.26 | 1.00 | 0.74 | non | oui | 2026-11-04T06:04:51.817315Z |
| deepsearchqa | search | 0.75 | 1.00 | 0.75 | non | oui | 2026-11-04T06:04:51.713300Z |
| gaia2 | general | 0.43 | 1.00 | 0.43 | non | oui | 2026-11-04T06:04:51.743083Z |
| gdpval-aa | general | 953.00 | 3 000.00 | 0.32 | non | oui | 2026-11-04T06:04:51.901471Z |
| gpqa | reasoning | 0.83 | 1.00 | 0.83 | non | oui | 2026-11-04T06:04:51.863267Z |
| humanity's-last-exam | reasoning | 0.22 | 1.00 | 0.22 | non | oui | 2026-11-04T06:04:51.873199Z |
| ifbench | instruction_following | 0.77 | 1.00 | 0.77 | non | oui | 2026-11-04T06:04:51.843029Z |
| mcp-atlas | reasoning | 0.76 | 1.00 | 0.76 | non | oui | 2026-11-04T06:04:51.703556Z |
| mmlu-prox | language | 0.80 | 1.00 | 0.80 | non | oui | 2026-11-04T06:04:51.942842Z |
| mmmu-pro | vision | 0.74 | 1.00 | 0.74 | non | oui | 2026-11-04T06:04:51.813073Z |
| omnidocbench-1.5 | vision | 0.76 | 1.00 | 0.76 | non | oui | 2026-11-04T06:04:51.803127Z |
| osworld-verified | multimodal | 0.66 | 1.00 | 0.66 | non | oui | 2026-11-04T06:04:51.753330Z |
| pinchbench | code | 0.91 | 1.00 | 0.91 | non | oui | 2026-11-04T06:04:51.927664Z |
| scicode | reasoning | 0.44 | 1.00 | 0.44 | non | oui | 2026-11-04T06:04:51.776855Z |
| screenspot-pro | vision | 0.75 | 1.00 | 0.75 | non | oui | 2026-11-04T06:04:51.792120Z |
| siren-agentdojo-attack-success | agents | 0.28 | 1.00 | 0.72 | non | oui | 2026-11-04T06:04:51.832507Z |
| siren-agentdojo-utility | agents | 0.94 | 1.00 | 0.94 | non | oui | 2026-11-04T06:04:51.839446Z |
| skillsbench | code | 0.44 | 1.00 | 0.44 | non | oui | 2026-11-04T06:04:51.750119Z |
| swe-bench-pro | reasoning | 0.51 | 1.00 | 0.51 | non | oui | 2026-11-04T06:04:51.758658Z |
| swe-bench-verified | reasoning | 0.76 | 1.00 | 0.76 | non | oui | 2026-11-04T06:04:51.763972Z |
| tau2-airline | chat | 0.78 | 1.00 | 0.78 | non | oui | 2026-11-04T06:04:51.905058Z |
| tau2-retail | communication | 0.87 | 1.00 | 0.87 | non | oui | 2026-11-04T06:04:51.913344Z |
| tau2-telecom | communication | 0.94 | 1.00 | 0.94 | non | oui | 2026-11-04T06:04:51.923243Z |
| tau3-banking | agents | 0.23 | 1.00 | 0.23 | non | oui | 2026-11-04T06:04:51.723884Z |
| terminal-bench-2.1 | reasoning | 0.52 | 1.00 | 0.52 | non | oui | 2026-11-04T06:04:51.772200Z |
| wildclawbench | agents | 0.48 | 1.00 | 0.48 | non | oui | 2026-11-04T06:04:51.733122Z |
Classements
| Catégorie | Rang | Score | Note conservatrice | Benchmarks évalués | Méthode | Daté du | Provenance |
|---|---|---|---|---|---|---|---|
| chat | 7 | 0.80 | 28.63 | 1 | trueskill | 2026-11-04T08:08:43.789169Z | rankings |
| agents | 94 | — | 16.02 | 8 | trueskill | 2026-10-19 | model_history |
| code | 96 | — | 21.64 | 6 | trueskill | 2026-10-19 | model_history |
Données
Dates de curation (récupération depuis l'API) : models 2026-11-04 · benchmarks 2026-11-04 · scores 2026-11-04 · updates 2026-11-04 · account 2026-11-04 · model_history 2026-11-04 · rankings 2026-11-04