Parameters
Undisclosed
Context Length
500K
Modality
Multimodal
Architecture
Undisclosed
License
Proprietary
Release Date
21 Sept 2026
Knowledge Cutoff
-
API Pricing (per 1M)
Input: $2.00 · Output: $6.00
Rank
#47
| Benchmark | Score | Rank |
|---|---|---|
Mathematics | max 0.96 | 13 |
General | max 0.77 | 23 |
LiveBench Average | max 0.77 | 24 |
Web Development | max 1640 | 26 |
Data Analysis | max 0.77 | 27 |
Agent Arena | max 0.02 | 28 |
Agentic Coding | max 0.54 | 30 |
Coding | max 0.77 | 39 |
Reasoning | max 0.83 | 45 |
Intelligence Index | max 0.35 high 0.35 low 0.31 | 69 70 89 |
General Text | max 1443 | 80 |
Overall Rank
#47
Coding Rank
#52
Grok 4.7 is xAI's flagship frontier model optimized for coding, agentic tool workflows, and rigorous knowledge work. It provides self-verification capabilities and strong long-context reasoning over a 500k-token window.
Architecture specifications are undisclosed for proprietary models.
xAI's frontier intelligence models trained with reinforcement learning at unprecedented scale using the 200,000 GPU Colossus cluster. Grok 4 series demonstrates state-of-the-art performance in reasoning, coding, and multimodal understanding with native tool use capabilities. Features real-time search integration across X and the web, advanced reasoning through scaled RL training, and industry-leading performance on academic benchmarks. Designed for both immediate responses and extended thinking modes with vision capabilities.