Muse Spark
Muse Spark is the first model in the Muse family developed by Meta Superintelligence Labs. It is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration. It features a Contemplating mode that orchestrates multiple agents reasoning in parallel. It demonstrates competitive performance in multimodal perception, reasoning, health, and agentic tasks, with Contemplating mode achieving 58% on Humanity's Last Exam and 38% on FrontierScience Research.
Identité
| Organisation | Meta |
|---|---|
| Pays | — |
| Famille | — |
| Type | llm |
| Date de sortie | 2026-04-08 |
| Licence | proprietary |
| Licence commerciale | non |
| Licence open source | non |
| Poids ouverts | non |
| Auto-hébergement (DGX Spark) | non — poids fermés |
| Modalités | — |
| Fenêtre de contexte | 1 048 576 |
| KV cache / token (Q8) | — |
| Paramètres totaux | — |
| Paramètres actifs (session) | — |
| Source des paramètres actifs | — |
| Curation des paramètres actifs | — |
| Prix par session | — |
| 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) | — |
| Tokens d'entraînement | — |
| Coupure de connaissance | — |
| Créé le | 2026-05-07T16:53:31.155602Z |
| Mis à jour le | 2026-11-04T06:04:51.963286Z |
| Source | LLM Stats |
| URL | https://llm-stats.com/models/muse-spark |
Prix par fournisseur
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 « — ».
Aucune donnée.
Souveraineté qualifiée par offre (modèle × fournisseur) ; survolez le badge pour les critères.
SouverainetéScores par catégorie (top_scores)
| Catégorie | Score |
|---|---|
| image_to_text | 0.70 |
| math | 0.60 |
| multimodal | 0.70 |
| physics | 0.90 |
| reasoning | 0.70 |
| science | 0.40 |
| search | 0.70 |
| spatial_reasoning | 0.60 |
| frontend_development | 0.80 |
| general | 0.70 |
| grounding | 0.80 |
| healthcare | 0.60 |
| agents | 0.60 |
| biology | 0.90 |
| chemistry | 0.90 |
| code | 0.70 |
| communication | 0.90 |
| tool_calling | 0.80 |
| vision | 0.70 |
Scores par benchmark
| Benchmark | Catégorie | Score | Max | Normalisé (API) | Vérifié | Auto-déclaré | Daté du |
|---|---|---|---|---|---|---|---|
| arc-agi-v2 | reasoning | 0.42 | 1.00 | 0.42 | non | oui | 2026-11-04T06:04:52.016961Z |
| charxiv-r | reasoning | 0.86 | 1.00 | 0.86 | non | oui | 2026-11-04T06:04:51.971188Z |
| deepsearchqa | search | 0.75 | 1.00 | 0.75 | non | oui | 2026-11-04T06:04:52.053009Z |
| erqa | vision | 0.65 | 1.00 | 0.65 | non | oui | 2026-11-04T06:04:51.977903Z |
| frontierscience-research | reasoning | 0.38 | 1.00 | 0.38 | non | oui | 2026-11-04T06:04:52.092992Z |
| gpqa | reasoning | 0.90 | 1.00 | 0.90 | non | oui | 2026-11-04T06:04:52.023325Z |
| healthbench-hard | healthcare | 0.43 | 1.00 | 0.43 | non | oui | 2026-11-04T06:04:52.033320Z |
| humanity's-last-exam | reasoning | 0.58 | 1.00 | 0.58 | non | oui | 2026-11-04T06:04:52.077939Z |
| ipho-2025 | reasoning | 0.83 | 1.00 | 0.83 | non | oui | 2026-11-04T06:04:52.083184Z |
| livecodebench-pro | reasoning | 0.80 | 3 000.00 | 0.80 | non | oui | 2026-11-04T06:04:52.028663Z |
| medxpertqa | healthcare | 0.78 | 1.00 | 0.78 | non | oui | 2026-11-04T06:04:52.044394Z |
| mmmu-pro | vision | 0.80 | 1.00 | 0.80 | non | oui | 2026-11-04T06:04:51.974522Z |
| screenspot-pro | vision | 0.84 | 1.00 | 0.84 | non | oui | 2026-11-04T06:04:51.985265Z |
| simplevqa | vision | 0.71 | 100.00 | 0.71 | non | oui | 2026-11-04T06:04:51.981332Z |
| swe-bench-pro | reasoning | 0.52 | 1.00 | 0.52 | non | oui | 2026-11-04T06:04:52.059833Z |
| swe-bench-verified | reasoning | 0.77 | 1.00 | 0.77 | non | oui | 2026-11-04T06:04:52.056558Z |
| tau2-telecom | communication | 0.92 | 1.00 | 0.92 | non | oui | 2026-11-04T06:04:52.067090Z |
| terminal-bench-2 | reasoning | 0.59 | 1.00 | 0.59 | non | oui | 2026-11-04T06:04:52.062988Z |
| zerobench | vision | 0.33 | 100.00 | 0.33 | non | oui | 2026-11-04T06:04:51.992975Z |
Classements
| Catégorie | Rang | Score | Note conservatrice | Benchmarks évalués | Méthode | Daté du | Provenance |
|---|---|---|---|---|---|---|---|
| math | 14 | 0.60 | 40.23 | 1 | trueskill | 2026-11-04T08:08:42.442591Z | rankings |
| multimodal | 30 | 0.70 | 31.75 | 6 | trueskill | 2026-11-04T08:08:46.488887Z | rankings |
| agents | 105 | — | 14.09 | 3 | trueskill | 2026-10-19 | model_history |
| code | 90 | — | 22.75 | 4 | 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