Grok 4 Heavy
xAI's premium variant with parallel-agent orchestration for the most demanding tasks
Grok 4 Heavy
SpaceXAI • July 2025
Training Data
Up to July 2025
Grok 4 Heavy
July 2025
Parameters
~2.4 trillion (estimated)
Training Method
Advanced RL with Agent Orchestration
Context Window
256,000 tokens
Knowledge Cutoff
Real-time via X
Key Features
Parallel Agent Orchestration • Premium Performance • Extended Thinking • Heavy Compute
Capabilities
Reasoning: Maximum
Complex Tasks: Outstanding
Agent Coordination: Outstanding
What's New in This Version
Premium variant using parallel-agent orchestration for maximum performance on complex tasks
xAI's premium variant with parallel-agent orchestration for the most demanding tasks
What's New in This Version
Premium variant using parallel-agent orchestration for maximum performance on complex tasks
Technical Specifications
Key Features
Capabilities
Other SpaceXAI Models
Explore more models from SpaceXAI
Grok 4.7
SpaceXAI's best model to date — a larger Grok trained on a longer, harder reinforcement-learning run plus SpaceX engineering data, at Grok 4.6's price and speed
Grok 4.6
SpaceXAI's frontier model for coding, agentic tasks, and knowledge work, focused on long-running agents and ambitious interactive and visual projects
Grok 4.5
SpaceXAI's coding and agentic-work flagship on the new V9 foundation — pitched as Opus-class capability but faster, more token-efficient, and lower cost
