In a post on X, Elon Musk announced that Grok 4.6 will be a 1.5-trillion-parameter model featuring significantly improved supervised fine-tuning (SFT) and reinforcement learning (RL), with an August target for launch. Grok 4.7 is planned as a 2.1-trillion-parameter model, which Musk stated "will be better than 4.6 in every way, except slightly slower to serve, albeit with even better token efficiency." The announcement follows the release of Grok 4.5 earlier this month and reflects xAI's rapid development cycle as competition intensifies among leading AI developers, including OpenAI, Anthropic and emerging Chinese firms. Musk's timeline and technical specifications have not been independently verified by xAI beyond his public statements, and the planned release dates could change.