xAI's new flagship model — a fast, lower-cost reasoning model for coding, knowledge work, conversation, and multimodal understanding.
Last verified · 2026-07-08 · by Moe Ameen
Grok 4.5 is the newest flagship model from xAI (the Musk-founded AI company now branded SpaceXAI). It is a general-purpose reasoning model — built for coding, agentic tasks, and knowledge work, with multimodal understanding (it reads text and images) and conversational use. It entered private beta with SpaceX and Tesla teams on June 28, 2026, and xAI set a wider public launch for July 9, 2026; it is already reachable through the xAI/SpaceXAI console and API, inside Grok Build, and in Cursor on all plans. At launch it was not yet available in the EU.
Elon Musk has described it as an "Opus-class" model — meaning roughly on par with what was recently Anthropic's top Claude family — but faster, more token-efficient, and lower cost. A chart xAI published with the announcement claims Grok 4.5 outperforms Opus 4.8 on several benchmarks; treat those as the company's own numbers until independent evaluations land. Reporting ties the model to xAI's new V9 foundation, said to be around 1.5 trillion parameters and supplemented with training data from the Cursor coding platform, though xAI has not published a full technical report and the weights are closed.
API pricing is $2.00 per million input tokens and $6.00 per million output tokens, with configurable reasoning effort (low, medium, or high; high by default) so you can trade depth for speed and cost. The model accepts text and image input and returns text.
One thing to be clear about for creators: Grok 4.5 is a model, not a content app. It can reason, draft, and analyze — write a script, outline a thread, summarize a document, read a screenshot — but it does not generate images or video, does not design or caption anything, and does not publish to any platform. It is the thinking-and-writing layer, not the production-and-distribution one.
Grok 4.5 is a strong place to think out loud and draft. Where it stops is the moment those words need to become a finished, on-brand post: it renders no pixels, holds no brand template, and publishes nowhere. That is the exact handoff Kompozy is built for. The practical workflow is ideation-to-production: brainstorm angles and rough copy in Grok 4.5, then drop the best of it — a script, an outline, a paragraph of raw notes — into Kompozy as a source. Kompozy fans that single input into finished formats it generates itself: a HeyGen persona/avatar short with auto-captions, a brand-exact carousel through HyperFrames, quote cards and photo posts, a blog article, and a newsletter — all in your voice via the Persona Brief, then scheduled and published across the nine connected social platforms plus email and blog from one queue.
The distinction that matters: Grok 4.5's own multimodal skill is understanding — reading an image or a document as input — not creating media. Kompozy is where the creating and shipping happens. So you use Grok 4.5 for the part it is genuinely good at (fast reasoning and drafting), and Kompozy for everything downstream that a raw model cannot do — the avatar video, the design, the captions, the platform-native sizing, the schedule. Note that Kompozy runs its own generation on managed Claude and OpenAI models, so you are not wiring Grok into it; you are pasting Grok's drafts in as raw material and letting the engine take them the rest of the way.
Grok 4.5 is xAI's newest flagship model — a general-purpose reasoning model for coding, agentic tasks, knowledge work, and conversation, with multimodal understanding of text and images. It entered private beta on June 28, 2026, with a wider public launch set for July 9, 2026, and is reachable via the xAI/SpaceXAI API and console, Grok Build, and Cursor.
Via the xAI API, Grok 4.5 is priced at $2.00 per million input tokens and $6.00 per million output tokens, with configurable reasoning effort (low, medium, or high; high by default). Consumer access may also ride xAI subscription tiers.
Elon Musk calls it "Opus-class" — roughly comparable but faster, more token-efficient, and lower cost — and xAI published a chart claiming it beats Opus 4.8 on several benchmarks. Those are xAI's own figures; wait for independent evaluations before treating the comparison as settled.
No. Grok 4.5 reasons over text and images and writes text; it generates no images, video, or audio and publishes nothing. To turn its drafts into finished, scheduled posts, pair it with a content engine like Kompozy that generates the media and publishes across platforms.
Use it as the drafting and ideation layer: brainstorm angles, write scripts, and summarize source material. Then paste those drafts into Kompozy, which generates persona video, carousels, images, blogs, and newsletters in your brand voice and publishes them to nine platforms plus email and blog.