Grok 4.6

Grok 4.6 is SpaceXAI's current flagship model for coding, agentic tool use, knowledge work, and interactive or visual tasks. The official model slug is grok-4.6. SpaceXAI documents a 500K-token context window, configurable reasoning, text and image input with text output, and standard API pricing of $2 per million input tokens and $6 per million output tokens.

Official sources: SpaceXAI models and pricing and Introducing Grok 4.6. Independent provider data should be treated separately from SpaceXAI's launch claims.

SpaceXAI positions Grok 4.6 as the successor to Grok 4.5, with a particular focus on long-running agents, coding, knowledge work, and ambitious interactive or visual workflows. It supports configurable reasoning so teams can trade latency and token use for deeper work.

• Repository-scale coding and debugging
• Long-running tool-connected agents
• STEM and technical knowledge work
• Image-aware product and interface analysis
• Workflows using web, X, code execution, or remote tools

Provider benchmark claims do not guarantee success on a specific repository or agent harness. Long-context requests cost more to run in some routes, and tool permissions, data handling, and provider availability vary. Test the exact workflow with an explicit rubric before production use.