xAI

Grok 4

Multimodal

by xAI

+
+
+
+
About

Grok 4, released by xAI on July 10, 2025, was built around first-principles reasoning — designed to approach problems through structured logical derivation rather than surface-level pattern matching. xAI positioned it specifically for advanced scientific and mathematical reasoning, and it was noted in the research community for its performance on frontier math benchmarks that typically distinguish dedicated reasoning models from generalist ones.

+
+
+
+
Pricing Range
Input (per 1M)$5.00 -$5.00
Output (per 1M)$25.00 -$25.00
Providers1
+
+
+
+
Timeline
ReleasedJul 9, 2025
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown

Overall Performance

4 benchmarks
Average Score
47.9%
Best Score
87.5%
High Performers (80%+)
1

Performance Metrics

Max Context Window
278.5K

Top Categories

Science
87.5%
Finance
53.5%
Reasoning
29.4%
Agents
21.1%
+
+
+
+
All Benchmark Results for Grok 4
Complete list of benchmark scores with detailed information
GPQA Diamond
Science
87.50
87.5%
Unverified
Finance Agent
Finance
53.51
53.5%
Unverified
ARC-AGI-2
Reasoning
29.40
29.4%
Unverified
GDPVal
Agents
21.10
21.1%
Unverified