Musk Says Grok 4.8 Is 'Not as Good' as Anthropic's Claude Opus 5.5

Asked how the two compare, Musk answered: "Not as good. Opus 5.5 is a masterpiece. But Grok 5 should exceed Opus 5.5 and maybe that happens in 2 or 3 months, with continued acceleration thereafter."

The founders of frontier labs rarely concede a lead to a competitor in public, and Musk's comment comes days before xAI is due to release Grok 4.8. It also puts the gap between the two companies in Musk's own words while Anthropic is pitching its initial public offering to investors. Musk did not publish any scores to show how far Grok 4.8 trails, and xAI has released no benchmarks for the model.

Musk has said Grok 4.8 will start a limited rollout to Grok bot users this week, after delays he has attributed to a rebuild of xAI's training software. He has described the model as not yet incredible but built on the right foundations. He has a long record of missing his own product timelines, so the two-to-three-month window for Grok 5 is a hedged estimate rather than a commitment.

The remark lands during a busy week for Anthropic. On Monday it released Claude Fable 5.2 and Claude Mythos 5.2, its first so-called frontier models since chief executive Dario Amodei called in September for the industry to slow the pace of AI development. On Tuesday it set a price range of $70 to $78 a share for its Nasdaq listing, which could value it at up to $1.405 trillion on a fully diluted basis, and launched spreadsheet and whiteboard tools for Claude.

The first independent comparison will be possible once Grok 4.8 reaches users and outside evaluators can test it against Anthropic's models.