Grok 4.6 by SpaceXAI: flagship code model with 500K context, configurable reasoning effort, minimal hallucinations, and no realtime access without search
Capabilities, design details, and architectural traits
Grok 4.6 is xAI's flagship model for code and general-purpose tasks, defined by three stated pillars: agentic tool calling, minimal hallucinations, and configurable reasoning.
| Trait | Detail |
|---|---|
| Configurable reasoning | Reasoning depth adjustable via reasoning_effort parameter |
| Context window | 500,000 tokens |
| Minimal hallucinations | Explicit design goal stated in model documentation |
Grok 4.6 reuses the same V9 foundation as its predecessor Grok 4.5, with capability gains delivered through improved supervised fine-tuning and reinforcement learning rather than increased parameter scale. The stated goal is matching or exceeding competing frontier models while preserving the inference speed and token efficiency of Grok 4.5. Supplemental training incorporates SpaceX engineering data, excluding ITAR-restricted material.
Independent evaluations · Artificial Analysis
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | SpaceXAI | Anthropic | Anthropic |
| Release Date | August 12, 2026 | July 24, 2026 | June 9, 2026 |
| Knowledge Cutoff | Feb 2026 | May 2026 | - |
| Context & Limits | |||
| Context Window | 500K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $2 Best Input Pricing | $5 | $10 |
| Output Pricing | $6 Best Output Pricing | $25 | $50 |
| Modalities | |||
| Inputs | text | textimage | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 60.9 | 63.1 Best Intelligence Index | 62.1 |
| Coding Index | 76.8 | 78.0 Best Coding Index | 76.5 |
| Agentic Index | 58.7 | 59.2 Best Agentic Index | 56.6 |
Graduate-level reasoning and expert Q&A evaluation.
Extremely difficult logical reasoning and knowledge.
Logical reasoning over long context windows.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
