SpaceXAI: Grok 4.5

Grok 4.5, listed by x-ai under the name SpaceXAI: Grok 4.5, accepts text, image, and file inputs and offers a 500,000-token context window with up to 450,000 completion tokens. It supports tool calling and reasoning, which suits agentic workflows and long multi-step tasks. Structured output support is not confirmed in the available data, and its weights status is unspecified, so teams that depend on strict schema adherence or self-hosting should verify both before committing. At $2.00 per million input tokens and $6.00 per million output tokens, it is a paid model whose large output allowance matters most for long-document generation and extended reasoning runs. Its blended benchmark score of 65.0 rests on a single independent benchmark, so its standing against rivals is largely unproven. Shortlist it if your work needs very long context plus tools, and plan to validate quality on your own tasks.