Grok 4.6 Makes Agent Tasks Costlier

Grok 4.6 highlights why token pricing alone is a poor proxy for agent economics.

• –Reasoning tokens are billed as output, so longer internal work directly increases costs
• –Cached input rises from $0.30 to $0.50 per million tokens versus Grok 4.5
• –Long-context requests above 200,000 tokens double Grok 4.6 rates to $4 input, $1 cached, and $12 output per million
• –More steps may improve task completion, but developers need per-task cost and success-rate metrics
• –Budget controls, step limits, prompt compaction, and model routing become essential for production agents