Saltar al contenido principal

Grok 4

SpaceXAIGrokProprietary

Descripción

Grok 4, announced by xAI in summer 2025, represents a major leap in AI capabilities, described as 'the smartest AI in the world.' Built on version 6 of xAI's foundation model, it uses 100x more training compute than Grok 2 and 10x more reinforcement learning compute than Grok 3. The model achieves PhD-level performance across all academic disciplines simultaneously, scoring perfect on standardized tests like the SAT and near-perfect on graduate exams like the GRE. Unlike Grok 3, tool usage is built into the training process rather than relying on generalization. Trained using 200,000 GPUs, Grok 4 excels at complex reasoning, mathematical problem-solving, and coding tasks, though it has acknowledged weaknesses in multimodal capabilities that are being addressed in the next version.

Fecha de lanzamiento
2025-07-10
Parámetros
Longitud del contexto
Modalidades
image, text

Radar de capacidades

49
general
74
coding
94
reasoning
61
science
88
agents
80
multimodal

Rankings

Dominio#PosiciónPuntuaciónFuente
Ranking de codificación75
75.0
AA
Ranking general115
67.0
AA
Ciencia81
72.0
AA

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Biology

GPQANYU + Cohere + Anthropic (2023)87.5%Aut.

Code

LiveCodeBench79.0%Aut.

Math

AIME 202591.7%Aut.
HMMT2590.0%Aut.
Humanity's Last Exam40.0%Aut.
USAMO2537.5%Aut.

Reasoning

ARC-AGI v215.9%Aut.

Índices de evaluación AA

(Artificial Analysis)
Math Index(Artificial Analysis)
92.7
Intelligence Index(Artificial Analysis)
34.1
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
1.0
Aime(MAA (Mathematical Association of America))
0.9
Aime 25(MAA (Mathematical Association of America))
0.9
Gpqa(NYU + Cohere + Anthropic (2023))
0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.9
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.8
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.7
Lcr(Artificial Analysis)
0.7
Ifbench(Google Research (2023))
0.5
Scicode(UIUC + Argonne National Lab (2024))
0.5
Terminalbench Hard(Stanford × Laude Institute (2026))
0.4
Hle(Center for AI Safety + Scale AI (2025))
0.3

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Physics
90
Biology
90
Chemistry
90
Code
80
Math
60
Reasoning
60
General
60
Vision
30
Spatial Reasoning
20

Precios

Precio de entrada$3 / 1M tokens
Precio de salida$15 / 1M tokens
Precio mixto (3:1)$6 / 1M tokens

Velocidad

Tokens/seg0.0
Retraso del primer token0.00s
Tiempo hasta la respuesta0.00s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

3 proveedores

Más barato: SpaceXAIMás caro: LLM Gateway
ProveedorEntradaSalida
1SpaceXAIPRINCIPAL
$3
$15
2Helicone
$3
$15
3LLM Gateway
$3
$15

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas