**Grok 4.6, xAI’s 1.5-trillion-parameter successor to Grok 4.5, shipped in early August 2026 with post-training upgrades targeting coding, agentic workflows, and engineering tasks.** Prior Grok 4 variants posted competitive HLE results, including up to 50% on multi-agent “Heavy” configurations, though public leaderboards remain led by Anthropic’s Claude Fable 5.1 and Opus 5 at 64.5–65%. The market reflects trader focus on whether the latest Grok release will promptly appear on the official HLE leaderboard at frontier-comparable accuracy, given xAI’s pattern of benchmark emphasis and the benchmark’s rapid score climb from single-digit percentages in early 2025 to the mid-40s–60s% range today. Key near-term catalysts include any verified HLE submission, a potential Grok 4.7 follow-up, and ongoing competition in reasoning and expert-knowledge evaluations.