Passer au contenu principal

Grok 3

xAIGrokProprietary

Description

Grok 3, launched by xAI on February 17, 2025, is an advanced AI model with significantly enhanced capabilities compared to Grok 2, boasting an order of magnitude increase in performance. Trained on a vast dataset that includes legal documents among others, and utilizing a massive compute infrastructure with around 200,000 GPUs in a Memphis data center, Grok 3's training used ten times more compute than its predecessor. It features specialized models like Grok 3 Reasoning and Grok 3 Mini Reasoning for complex problem-solving, and it excels in benchmarks like AIME for mathematics and GPQA for PhD-level science.

Date de sortie
2025-02-19
Paramètres
Longueur du contexte
131K
Modalités
image, text

Radar de capacités

39
general
29
coding
57
reasoning
45
scienceest.
0
agents
80
multimodal

Science utilise un proxy de raisonnement lorsque les benchmarks scientifiques dédiés ne sont pas disponibles.

Classements

Domaine#RangScoreSource
Code Ranking201
41.0
AA
General Ranking186
52.0
AA
Math Reasoning150
60.0
AA
Science210
47.0
AA

Scores de benchmarks (LLM Stats)

Biology

GPQA84.6%Aut.

Code

LiveCodeBench79.4%Aut.

General

MMMU78.0%Aut.

Math

AIME 202593.3%Aut.
AIME 202493.3%Aut.

Indices d'évaluation AA

Math Index
58.0
Intelligence Index
25.2
Coding Index
19.8
Math 500
0.9
Mmlu Pro
0.8
Gpqa
0.7
Aime 25
0.6
Lcr
0.5
Tau2
0.5
Ifbench
0.5
Livecodebench
0.4
Scicode
0.4
Aime
0.3
Terminalbench Hard
0.1
Hle
0.1

Scores par catégorie LLM Stats

Math
90
Reasoning
90
Vision
80
Biology
80
Chemistry
80
Code
80
General
80
Healthcare
80
Multimodal
80
Physics
80

Tarification

Prix d'entrée$3 / 1M tokens
Prix de sortie$15 / 1M tokens
Prix mixte (3:1)$6 / 1M tokens

Vitesse

Tokens/sec43.6 tokens/s
Délai du premier token0.59s
Temps de réponse0.59s

Fournisseurs disponibles

(Unités internes LS)
FournisseurPrix d'entréePrix de sortie
xAI3.0M15.0M

Sources externes