Rankings
Overall Rank
-
Coding Rank
-
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
Attention
Attention Structure
Multi-Head Attention
Attention Heads
-
Key-Value Heads
-
Attention Head Dimension
-
Position Embedding
ROPE
RoPE Theta
-
Sliding Window Attention
-
Sliding Window Size
-
Sliding Window Ratio
-
Linear Attention
-
Linear Attention Ratio
-
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Hidden Dimension Size
-
Number of Layers
-
FFN Intermediate Size (Dense)
-
Multi-Token Prediction Heads
-
Tokenizer
Vocabulary Size
-
Mixture of Experts
Total Expert Parameters
-
Number of Experts
-
Active Experts
-
Shared Experts
-
FFN Intermediate Size (per Expert)
-
Dense Layers Before MoE
-
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.