Nemotron 3 Nano (30B A3B)
Nemotron 3 Nano is a 31.6B hybrid MoE model optimized for fast, long‑context agentic reasoning. It mixes Mamba‑2 and Transformer layers with a sparse MoE router (~3.6B active params per token) to deliver up to 4× higher throughput than Nemotron 2 and strong accuracy across math, coding, and tools. It supports a 1M‑token context window, offers Reasoning ON/OFF and a thinking‑budget to control costs, and ships with open weights, data, and RL tooling (NeMo Gym/RL). Released Dec 15, 2025 under the NVIDIA Open Model License, it’s built as the efficient backbone for multi‑agent systems at scale.
Identité
| Organisation | NVIDIA |
|---|---|
| Pays | — |
| Famille | — |
| Type | llm |
| Date de sortie | 2025-12-15 |
| Licence | nvidia open model license agreement |
| Licence commerciale | oui |
| Licence open source | non |
| Poids ouverts | oui |
| Auto-hébergement (DGX Spark) | oui — ≤ 107 Md (Q8/FP8) |
| Modalités | text |
| Fenêtre de contexte | — |
| KV cache / token (Q8) | — |
| Paramètres totaux | 32 B |
| Paramètres actifs (session) | 3 B |
| Source des paramètres actifs | actifs : nom (heuristique) ; total : nom (heuristique) |
| Curation des paramètres actifs | 2026-11-04 |
| Prix par session | 0.09 $ (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) | 0.76 $ (est.) |
| Tokens d'entraînement | 25 000 000 000 000 |
| Coupure de connaissance | 2025-11-28 |
| Créé le | 2026-05-07T16:53:33.118394Z |
| Mis à jour le | 2026-11-04T06:04:58.410571Z |
| Source | LLM Stats |
| URL | https://llm-stats.com/models/nemotron-3-nano-30b-a3b |
IA Act — preuves fournisseur
| Critère | Note / 5 | Poids |
|---|---|---|
| Documentation technique | 2,5 | 5 % |
| Résumé des données d'entraînement | 0,0 | 35 % |
| Conformité droit d'auteur | 0,0 | 35 % |
| Code of Practice GPAI | 0,0 | 20 % |
| Risque systémique | 0,0 | 5 % |
Curation : 2026-10-23.
IA ActPrix par fournisseur
Statut du modèle (offres + auto-hébergement) : souverain — via Nebul (eurouter).
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 |
|---|---|---|---|---|---|---|---|---|---|
| Nebul | eurouter | souverain | — | — | — | 262 144 | — | — | — |
| DeepInfra | llm-stats | non | 0.05 $ | 0.20 $ | — | — | — | — | active |
Souveraineté qualifiée par offre (modèle × fournisseur) ; survolez le badge pour les critères.
SouverainetéScores par catégorie (top_scores)
| Catégorie | Score |
|---|---|
| chat | 0.40 |
| language | 0.70 |
| legal | 0.70 |
| math | 0.60 |
| physics | 0.50 |
| reasoning | 0.50 |
| finance | 0.70 |
| frontend_development | 0.40 |
| general | 0.50 |
| healthcare | 0.70 |
| agents | 0.10 |
| biology | 0.50 |
| chemistry | 0.50 |
| code | 0.30 |
| communication | 0.50 |
| creativity | 0.70 |
| tool_calling | 0.50 |
| vision | 0.20 |
| writing | 0.70 |
Scores par benchmark
| Benchmark | Catégorie | Score | Max | Normalisé (API) | Vérifié | Auto-déclaré | Daté du |
|---|---|---|---|---|---|---|---|
| aime-2025 | math | 0.99 | 1.00 | 0.99 | non | oui | 2026-11-04T06:04:58.420132Z |
| arena-hard-v2 | general | 0.68 | 1.00 | 0.68 | non | oui | 2026-11-04T06:04:58.471677Z |
| gpqa | reasoning | 0.75 | 1.00 | 0.75 | non | oui | 2026-11-04T06:04:58.426056Z |
| humanity's-last-exam | reasoning | 0.15 | 1.00 | 0.15 | non | oui | 2026-11-04T06:04:58.438824Z |
| livecodebench-v6 | reasoning | 0.68 | 1.00 | 0.68 | non | oui | 2026-11-04T06:04:58.429636Z |
| mmlu-pro | language | 0.78 | 1.00 | 0.78 | non | oui | 2026-11-04T06:04:58.413833Z |
| mmlu-prox | language | 0.59 | 1.00 | 0.59 | non | oui | 2026-11-04T06:04:58.477547Z |
| multichallenge | chat | 0.39 | 1.00 | 0.39 | non | oui | 2026-11-04T06:04:58.462449Z |
| scicode | reasoning | 0.33 | 1.00 | 0.33 | non | oui | 2026-11-04T06:04:58.432540Z |
| swe-bench-verified | reasoning | 0.39 | 1.00 | 0.39 | non | oui | 2026-11-04T06:04:58.445368Z |
| tau2-airline | chat | 0.48 | 1.00 | 0.48 | non | oui | 2026-11-04T06:04:58.448136Z |
| tau2-retail | communication | 0.57 | 1.00 | 0.57 | non | oui | 2026-11-04T06:04:58.450974Z |
| tau2-telecom | communication | 0.42 | 1.00 | 0.42 | non | oui | 2026-11-04T06:04:58.459146Z |
| terminal-bench | reasoning | 0.09 | 1.00 | 0.09 | non | oui | 2026-11-04T06:04:58.441748Z |
| wmt24++ | language | 0.86 | 1.00 | 0.86 | non | oui | 2026-11-04T06:04:58.480799Z |
Classements
| Catégorie | Rang | Score | Note conservatrice | Benchmarks évalués | Méthode | Daté du | Provenance |
|---|---|---|---|---|---|---|---|
| agents | 199 | — | -6.69 | 1 | trueskill | 2026-10-19 | model_history |
| code | 254 | — | 0.43 | 3 | 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