Grok 4.6
Active Parameters
Undisclosed
Context Length
500K
Modality
Multimodal
Architecture
Mixture of Experts (MoE)
License
Proprietary
Release Date
1 Aug 2026
Knowledge Cutoff
Feb 2026
API Pricing (per 1M)
Input: $2.00 · Output: $6.00
Evaluation Benchmarks
Rank
#7
| Benchmark | Score | Rank |
|---|---|---|
Agentic Index | max 0.53 high 0.54 medium 0.51 low 0.42 | 8 5 12 30 |
Reasoning | 0.91 | 6 |
Web Development | high 1625 | 7 |
Coding Index | max 0.76 high 0.77 medium 0.74 low 0.66 | 16 7 20 52 |
Intelligence Index | max 0.49 high 0.51 medium 0.48 low 0.42 | 12 8 14 42 |
General | 0.78 | 10 |
LiveBench Average | 0.78 | 10 |
Agentic Coding | 0.57 | 15 |
Mathematics | 0.93 | 15 |
Agent Arena | max 0.05 | 17 |
Data Analysis | 0.74 | 26 |
Coding | 0.77 | 28 |
General Text | high 1461 | 41 |
Rankings
Overall Rank
#7
Coding Rank
#8
About Grok 4.6
Grok 4.6 is xAI's flagship frontier multimodal intelligence model trained on the massive Colossus supercluster. Featuring native vision understanding, real-time X and web grounding, and extended reasoning capabilities, Grok 4.6 is designed for complex mathematical reasoning, enterprise coding, and high-autonomy agent workflows with a 500,000-token context window.
Technical Specifications
Architecture specifications are undisclosed for proprietary models.
Resources
About Grok 4
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.