Grok-3
xAI's advanced model trained with 10x more compute than Grok-2
Grok-3
SpaceXAI • February 2025
Training Data
Up to early 2025
Grok-3
February 2025
Parameters
~500 billion
Training Method
Advanced Reinforcement Learning
Context Window
1,000,000 tokens
Knowledge Cutoff
Real-time via X
Key Features
Enhanced Reasoning • Real-time Data • Think Mode • 1M Token Context
Capabilities
Reasoning: Outstanding
Current Events: Outstanding
Context: Outstanding
What's New in This Version
Major advancement in reasoning with 8x larger context window and Think mode
xAI's advanced model trained with 10x more compute than Grok-2
What's New in This Version
Major advancement in reasoning with 8x larger context window and Think mode
Technical Specifications
Key Features
Capabilities
Other SpaceXAI Models
Explore more models from SpaceXAI
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
Grok 4.3
xAI's reasoning flagship with always-on reasoning, native video input, 1M context, and significantly cheaper API pricing
Grok 4.20
xAI's flagship multi-agent model combining four specialized agents on a shared MoE backbone for industry-leading speed and lowest hallucination rates
