xAI Releases Grok 4.6, Reaching the Intelligence Frontier at Lower Cost

▪

xAI officially released Grok 4.6, its new flagship artificial intelligence model, achieving a score of 61 on the Artificial Analysis Intelligence Index. The release places xAI back at the intelligence frontier alongside OpenAI's GPT-5.6 Sol and close to Anthropic's leading models. Unlike previous generations that relied heavily on expanding base model size, Grok 4.6 achieves its performance gains primarily through advanced post-training passes, including refined supervised fine-tuning and intensive reinforcement learning on agentic tasks. It retains the 1.5 trillion parameter V9 foundation from Grok 4.5 while significantly boosting multi-turn execution and tool usage capabilities. Independent evaluations show Grok 4.6 scoring an Elo of 1,753 on GDPval-AA v2, reflecting strong competency in real-world agentic knowledge work. It also achieved 50.7 percent on the tau-cube banking evaluation and 88.4 percent on Terminal-Bench v2.1 for terminal-based software tasks. The model supports a 500k token context window and accepts text and image inputs. Despite closing the intelligence gap with top-tier competitors, xAI maintained headline pricing flat at $2.00 per million input tokens and $6.00 per million output tokens. This cost structure positions Grok 4.6 significantly below comparable frontier models such as Claude Opus 5 and GPT-5.6 Sol, which trade at higher price points per token. The model is now available through the developer API, Cursor, and Grok Build environments.

Prepared by Jonathan Pierce and reviewed by editorial team.

Media Bias

Articles Published:: 29
Right Leaning:: 0
Left Leaning:: 0
Neutral:: 29

Explain Framing

Emphasized competitive market shifts and cost pressures on big tech. Focused objectively on benchmark improvements, pricing structures, and independent metrics. Highlighted rapid technological scaling, American innovation leadership, and economic value.