xAI: Grok 4

xAI: Grok 4

x-ai · Released Jul 9, 2025
80
Our Score

Grok 4 is xAI's latest reasoning model with a 256k context window. It supports parallel tool calling, structured outputs, and both image and text inputs. Note that reasoning is not exposed, reasoning cannot be disabled, and the reasoning effort cannot be specified. Pricing increases once the total tokens in a given request is greater than 128k tokens.

$3.00 / 1M Input Price
$15.00 / 1M Output Price
256,000 tokens Context Window

Capabilities

Tool Use Function Calling Vision

Architecture

ModalityText + Image → Text
TokenizerGrok

Performance Indices

Source: Artificial Analysis

41.5 Intelligence Index
40.5 Coding Index
56.4 Agentic Index
92.7 Math Index

Benchmark Scores

Evaluations

GPQA Diamond 87.7%
Graduate-level scientific reasoning
HLE 23.9%
Humanity's Last Exam
MMLU Pro 86.6%
Multi-task language understanding
LiveCodeBench 81.9%
Live coding evaluation
SciCode 45.7%
Scientific computing
MATH 500 99%
Mathematical problem-solving
AIME 94.3%
Competition mathematics
AIME 2025 92.7%
Competition mathematics (2025)
IFBench 53.7%
Instruction following
LCR 68%
Long-context reasoning
TerminalBench Hard 37.9%
Agentic terminal tasks
τ²-Bench 74.9%
Conversational agent benchmark

Benchmark data from Artificial Analysis and Hugging Face

Model Information

OpenRouter ID x-ai/grok-4
Providerx-ai
Release Date July 9, 2025
Context Length256,000 tokens
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $3.00 $0.003000
Output $15.00 $0.015000

Live Performance

Live endpoint metrics — refreshed every 30 minutes.

100%
Avg Uptime
11,765ms
Best Latency (TTFT)
39 tok/s
Best Throughput
1/1
Active Endpoints
Available via: xAI

Leaderboard Categories