Jul 9, 2026 · AI News

xAI Unveils Grok 4.5: Context, Tools, and What’s Confirmed

Holographic robot reviewing a launch checklist for the Grok 4.5 AI model announcement

xAI has published a Grok 4.5 announcement page on its news site, confirming the model as the next entry in the Grok lineup. Separate reporting around the same launch period points to a 1 million token context window, an expanded tool suite, and $0.60 per million input tokens, though the source page was not directly accessible and several details remain unverified.

What xAI has confirmed

The announcement page itself is live at xAI’s news domain, confirming only that Grok 4.5 has been officially named as a successor in the Grok lineup.

What separate reporting points to

  • Reported context window: 1 million tokens.
  • Reported tools: web search, code execution, image generation.
  • Reported pricing: $0.60 per million input tokens.

What remains unverified

Because the source page could not be fetched, the model card, system prompt restrictions, benchmark results, and rollout timeline were not confirmed directly. Claims along those lines should be treated as unverified until xAI publishes reproducible documentation.

Why the reported figures matter

Token pricing and context window size tend to shape how teams budget for long-context workloads. If the reported figures hold, Grok 4.5 would enter a market where several frontier providers already advertise million-token contexts, so differentiators are more likely to live in tooling and latency than in raw scale.

FAQ

Has xAI officially announced Grok 4.5?

Yes. xAI has published a Grok 4.5 announcement page on its news site, confirming the name as the next entry in the Grok lineup.

What specs have been reported for Grok 4.5?

Separate reporting around the launch period points to a 1 million token context window, tools including web search, code execution, and image generation, and pricing of $0.60 per million input tokens.

Why are some Grok 4.5 details still unverified?

The xAI announcement page could not be fetched, so the model card, system prompt restrictions, benchmark results, and rollout timeline have not been independently confirmed.

Related coverage