Beta
xAI

Grok 4

por xAI · Lançado em 2025-07-09

47.8
pontuação média
$3.00/1M
Preço de entrada
$15.00/1M
Preço de saída
256K tokens (~128 books)
Janela de contexto
multimodal
Tipo

Tested on 17 benchmarks with 47.8% average. Top scores: Fiction.LiveBench (94.4%), OTIS Mock AIME 2024-2025 (84.0%), GPQA diamond (82.7%).

Pontuações de benchmark

BenchmarkCategoriaPontuaçãoBar
Fiction.LiveBenchknowledge94.4
OTIS Mock AIME 2024-2025math84.0
GPQA diamondknowledge82.7
Lech Mazur Writingknowledge80.7
Aider polyglotcoding79.6
SimpleBenchreasoning52.6
SimpleQA Verifiedknowledge47.9
DeepResearch Benchknowledge47.9
WeirdMLcoding45.7
GeoBenchknowledge45.0
Balrogknowledge43.6
Chess Puzzlesknowledge28.0
Terminal Benchcoding27.2
FrontierMath-2025-02-28-Privatemath19.7
ARC-AGI-2reasoning16.0
APEX-Agentsagentic15.2
FrontierMath-Tier-4-2025-07-01-Privatemath2.1

Modelos similares