Grok-4.1 Thinking

Fiche modèle · xAI · grok-4.1-thinking-2025-11-17

Grok 4.1 Thinking (code name: quasarflux) is the reasoning variant of Grok 4.1, bringing significant improvements to the real-world usability of Grok. The model is exceptionally capable in creative, emotional, and collaborative interactions. It is more perceptive to nuanced intent, compelling to speak with, and coherent in personality, while fully retaining the razor-sharp intelligence and reliability of its predecessors. Grok 4.1 Thinking uses thinking tokens for deeper reasoning, achieving state-of-the-art performance on blind human preference evaluations. The model features reduced hallucinations compared to previous versions, with significant improvements in factual accuracy for information-seeking prompts. To achieve this, xAI used large scale reinforcement learning infrastructure to optimize style, personality, helpfulness, and alignment, developing new methods that use frontier agentic reasoning models as reward models to autonomously evaluate and iterate on responses at scale. Grok 4.1 Thinking includes comprehensive safety mitigations including refusal training, input filters for restricted knowledge, and adversarial robustness measures. The model demonstrates strong refusal rates on harmful queries (7% answer rate on chat refusals, 14% on agentic refusals) and improved honesty training to reduce deception.

Identité

OrganisationxAI
Pays—
Famille—
Typellm
Date de sortie2025-11-17
Licenceproprietary
Licence commercialenon
Licence open sourcenon
Poids ouvertsnon
Auto-hébergement (DGX Spark)non — poids fermés
Modalitésimage|text
Fenêtre de contexte—
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 Players69.4×
Coût / tâche (Terminal-Bench)—
Tokens d'entraînement—
Coupure de connaissance—
Créé le2026-05-07T16:53:39.952604Z
Mis à jour le2026-11-04T06:05:13.338417Z
SourceLLM Stats
URLhttps://llm-stats.com/models/grok-4.1-thinking-2025-11-17

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.

Souveraineté

Scores par catégorie (top_scores)

CatégorieScore
safety0.40
general0.50
healthcare0.30
agents0.40
code0.40
creativity0.90
vision0.30
writing0.90

Scores par benchmark

BenchmarkCatégorieScoreMaxNormalisé (API)VérifiéAuto-déclaréDaté du
biolp-benchbiology0.371.000.37nonoui2026-11-04T06:05:13.375937Z
cloningscenariossafety0.461.000.46nonoui2026-11-04T06:05:13.385846Z
creative-writing-v3creativity0.861.000.86nonoui2026-11-04T06:05:13.356260Z
cybenchsafety0.391.000.39nonoui2026-11-04T06:05:13.388917Z
eq-benchchat1 586.002 000.000.79nonoui2026-11-04T06:05:13.350085Z
figqasafety0.341.000.34nonoui2026-11-04T06:05:13.382375Z
lmarena-textchat1 483.002 000.000.74nonoui2026-11-04T06:05:13.346627Z
masksafety0.511.000.51nonoui2026-11-04T06:05:13.392752Z
protocolqasafety0.791.000.79nonoui2026-11-04T06:05:13.378939Z
vctsafety0.611.000.61nonoui2026-11-04T06:05:13.371788Z
wmdpsafety0.841.000.84nonoui2026-11-04T06:05:13.368569Z

Classements

CatégorieRangScoreNote conservatriceBenchmarks évaluésMéthodeDaté duProvenance
agents184—1.331trueskill2026-10-19model_history
code250—1.011trueskill2026-10-19model_history

Lecture : Note conservatrice = borne basse TrueSkill (niveau estimé − marge d’incertitude) ; Benchmarks évalués = nombre de benchmarks de la catégorie pris en compte. À performances égales, un modèle évalué sur moins de benchmarks obtient une note plus basse : la note ne se déduit pas d’un seul score. Provenance : rankings = top 50 de /rankings ; model_history = historique du modèle (/models/{id}/history), utilisé pour les rangs au-delà du top 50.

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