Kimi K3
Kimi K3 is Moonshot AI's flagship open Mixture-of-Experts model for long-horizon coding, knowledge work, and reasoning. It has 2.8 trillion total parameters, activates 16 of 896 experts, and combines Kimi Delta Attention, Attention Residuals, and Stable LatentMoE. The model supports a 1M-token context window, native image and video understanding, tool calling, structured output, and always-on reasoning with max thinking effort at launch.
Identité
| Organisation | Moonshot AI |
|---|---|
| Pays | — |
| Famille | — |
| Type | llm |
| Date de sortie | 2026-07-16 |
| Licence | kimi k3 license |
| Licence commerciale | oui |
| Licence open source | non |
| Poids ouverts | oui |
| Auto-hébergement (DGX Spark) | non — 2.8e+03 Md > 107 Md |
| Modalités | image|text|video |
| Fenêtre de contexte | 1 048 576 |
| KV cache / token (Q8) | 2.29 Mo/token |
| Paramètres totaux | 2.8 T |
| Paramètres actifs (session) | 104 B |
| Source des paramètres actifs | web : https://huggingface.co/moonshotai/Kimi-K3 |
| Curation des paramètres actifs | 2026-11-04 |
| Prix par session | 5.70 $ (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) | 47.87 $ (est.) |
| Tokens d'entraînement | — |
| Coupure de connaissance | — |
| Créé le | 2026-07-17T02:50:10.663128Z |
| Mis à jour le | 2026-11-04T06:04:57.903524Z |
| Source | LLM Stats |
| URL | https://llm-stats.com/models/kimi-k3 |
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 Arkane (cortecs), Berget AI AB (cortecs, eurouter), GreenPT (eurouter), Lyceum Technology GmbH (eurouter), Nebius (cortecs), Nebul (eurouter), Tensorix (cortecs, 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 |
|---|---|---|---|---|---|---|---|---|---|
| Arkane | cortecs | souverain | 3.02 $ | 15.11 $ | — | 1 000 000 | — | — | — |
| Berget AI AB | cortecs, eurouter | souverain | 3.37 $ | 16.84 $ | mxfp4 | 1 000 000 / 1 048 576 | — | — | — |
| GreenPT | eurouter | souverain | — | — | — | 1 048 576 | — | — | — |
| Lyceum Technology GmbH | eurouter | souverain | — | — | — | 1 048 576 | — | — | — |
| Nebius | cortecs | souverain | 3.02 $ | 15.11 $ | fp4 | 1 024 000 | — | — | — |
| Nebul | eurouter | souverain | — | — | — | 1 048 576 | — | — | — |
| Tensorix | cortecs, eurouter | souverain | 3.02 $ | 15.11 $ | mxfp4 | 1 048 576 | — | — | — |
| Sference | cortecs | partiel | 3.02 $ | 15.11 $ | nvfp4 | 1 048 576 | — | — | — |
| DeepInfra | llm-stats | non | 2.85 $ | 14.25 $ | — | — | — | — | active |
| Fireworks AI | llm-stats | non | 3.00 $ | 15.00 $ | — | — | — | — | active |
| Moonshot AI | llm-stats | non | 3.00 $ | 15.00 $ | — | — | — | — | active |
| Novita | llm-stats | non | 3.00 $ | 15.00 $ | — | — | — | — | active |
| Together AI | llm-stats | non | 3.00 $ | 15.00 $ | — | — | — | — | active |
Souveraineté qualifiée par offre (modèle × fournisseur) ; survolez le badge pour les critères.
SouverainetéScores par catégorie (top_scores)
| Catégorie | Score |
|---|---|
| math | 0.80 |
| multimodal | 0.70 |
| physics | 0.90 |
| productivity | 516.30 |
| reasoning | 74.40 |
| search | 0.90 |
| general | 54.00 |
| agents | 82.10 |
| biology | 0.90 |
| chemistry | 0.90 |
| code | 0.70 |
| systems | 0.40 |
| tool_calling | 0.60 |
| vision | 0.70 |
Scores par benchmark
| Benchmark | Catégorie | Score | Max | Normalisé (API) | Vérifié | Auto-déclaré | Daté du |
|---|---|---|---|---|---|---|---|
| aa-briefcase | agents | 1 548.00 | 3 000.00 | 0.52 | non | non | 2026-11-04T06:04:58.013298Z |
| apex-agents | reasoning | 0.38 | 1.00 | 0.38 | non | oui | 2026-11-04T06:04:58.020127Z |
| automationbench | agents | 0.31 | 1.00 | 0.31 | non | oui | 2026-11-04T06:04:57.999079Z |
| babyvision | vision | 0.86 | 1.00 | 0.86 | non | oui | 2026-11-04T06:04:58.113366Z |
| browsecomp | reasoning | 0.91 | 1.00 | 0.91 | non | oui | 2026-11-04T06:04:57.971494Z |
| charxiv-r | reasoning | 0.91 | 1.00 | 0.91 | non | oui | 2026-11-04T06:04:58.091978Z |
| deck-bench | agents | 0.73 | 1.00 | 0.73 | non | oui | 2026-11-04T06:04:58.050002Z |
| deepsearchqa | search | 0.95 | 1.00 | 0.95 | non | oui | 2026-11-04T06:04:57.980089Z |
| deepswe | code | 0.68 | 1.00 | 0.68 | non | oui | 2026-11-04T06:04:57.913621Z |
| deepswe-1.1 | code | 0.69 | 1.00 | 0.69 | non | non | 2026-11-04T06:04:58.155141Z |
| frontierswe | code | 0.81 | 1.00 | 0.81 | non | oui | 2026-11-04T06:04:57.939648Z |
| gpqa | reasoning | 0.94 | 1.00 | 0.94 | non | oui | 2026-11-04T06:04:58.055527Z |
| humanity's-last-exam | reasoning | 0.56 | 1.00 | 0.56 | non | oui | 2026-11-04T06:04:58.063039Z |
| job-bench | agents | 0.53 | 1.00 | 0.53 | non | oui | 2026-11-04T06:04:58.009758Z |
| kimi-code-bench-v2 | agents | 0.73 | 1.00 | 0.73 | non | oui | 2026-11-04T06:04:57.966678Z |
| lifesci-diamond-2026-11 | healthcare | 0.09 | 1.00 | 0.09 | non | non | 2026-11-04T06:04:58.158677Z |
| mathvision | math | 0.98 | 1.00 | 0.98 | non | oui | 2026-11-04T06:04:58.107939Z |
| mcp-atlas | reasoning | 0.84 | 1.00 | 0.84 | non | oui | 2026-11-04T06:04:57.992977Z |
| mls-bench-lite | agents | 0.48 | 1.00 | 0.48 | non | oui | 2026-11-04T06:04:57.963301Z |
| mmmu-pro | vision | 0.82 | 1.00 | 0.82 | non | oui | 2026-11-04T06:04:58.067073Z |
| mmmu-pro-with-tools | vision | 0.83 | 1.00 | 0.83 | non | oui | 2026-11-04T06:04:58.076655Z |
| officeqa-pro | agents | 0.63 | 1.00 | 0.63 | non | oui | 2026-11-04T06:04:58.033183Z |
| omnidocbench | vision | 0.91 | 1.00 | 0.91 | non | oui | 2026-11-04T06:04:58.140099Z |
| perceptionbench | vision | 0.58 | 1.00 | 0.58 | non | oui | 2026-11-04T06:04:58.145809Z |
| posttrainbench | agents | 0.37 | 1.00 | 0.37 | non | oui | 2026-11-04T06:04:57.953069Z |
| program-bench | agents | 0.78 | 1.00 | 0.78 | non | oui | 2026-11-04T06:04:57.919876Z |
| spreadsheetbench-2 | agents | 0.35 | 1.00 | 0.35 | non | oui | 2026-11-04T06:04:58.039766Z |
| swe-marathon | code | 0.42 | 1.00 | 0.42 | non | oui | 2026-11-04T06:04:57.949160Z |
| terminal-bench-2.1 | reasoning | 0.88 | 1.00 | 0.88 | non | oui | 2026-11-04T06:04:57.931099Z |
| toolathlon | agents | 0.73 | 1.00 | 0.73 | non | oui | 2026-11-04T06:04:57.989052Z |
| worldvqa | vision | 0.51 | 1.00 | 0.51 | non | oui | 2026-11-04T06:04:58.133239Z |
| zerobench | vision | 0.41 | 100.00 | 0.41 | non | oui | 2026-11-04T06:04:58.126394Z |
Classements
| Catégorie | Rang | Score | Note conservatrice | Benchmarks évalués | Méthode | Daté du | Provenance |
|---|---|---|---|---|---|---|---|
| agents | 20 | 82.10 | 37.29 | 19 | trueskill | 2026-11-04T08:08:47.839150Z | rankings |
| code | 14 | 0.70 | 40.72 | 10 | trueskill | 2026-11-04T08:08:39.742311Z | rankings |
| math | 7 | 0.80 | 42.05 | 2 | trueskill | 2026-11-04T08:08:42.442591Z | rankings |
| multimodal | 9 | 0.70 | 36.64 | 8 | trueskill | 2026-11-04T08:08:46.488887Z | rankings |
| reasoning | 12 | 74.40 | 50.34 | 21 | trueskill | 2026-11-04T08:08:41.093494Z | rankings |
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