Grok 4
Current · Released: July 15, 2025
- Context window
- 256.000 tokens
- Max output
- 32.768 tokens
- Price input / 1M tokens
- $3.00
- Price output / 1M tokens
- $15.00
Capabilities
- Thinking
- Tool use
- Real-time search
Benchmark scores
- MMMU
- 75.0
- MMLU-Pro
- 87.0
- AIME 2024
- 94.0
- GPQA Diamond
- 87.5
- Humanity's Last Exam
- 25.4
Notes
Reasoning-heavy model with built-in X/Twitter search. Source: openrouter.ai.