The model behind the Grok Build CLI reached public beta on the xAI API at $1 per million input tokens and $2 per million output, with a 256k context window.
2026-06-25 · by Moe Ameen
xAI made Grok Build 0.1 — the coding model that powers its Grok Build CLI — available to developers directly through the xAI API in public beta in late May 2026, after a brief early-access window earlier that month. The model is exposed under the slug `grok-build-0.1`, and the move lets developers build on the same model that drives the CLI without a SuperGrok or X Premium+ subscription. xAI describes it as a model trained specifically for agentic coding: web development, debugging, and tool use via the Model Context Protocol.
The API pricing is $1.00 per million input tokens and $2.00 per million output tokens, with cached input billed at $0.20 per million — an 80% cache discount. It carries a 256k-token context window and accepts text and image input. xAI emphasizes speed: on Artificial Analysis's independent test of the June 2026 ("0616") snapshot, the model runs at roughly 104 output tokens per second with a time-to-first-token near half a second, while posting an Intelligence Index of 40, above the median for its price tier. Parameter count is undisclosed and the weights are closed.
The Grok Build CLI around the model uses agentic patterns now common in this category — plan-first execution, parallel sub-agents that can each run in their own Git worktree, and native MCP support. That places Grok Build alongside Anthropic's Claude Code and OpenAI's Codex CLI in the terminal-coding-agent race. Some third-party reports cite a SWE-Bench Verified score in the low 70s; treat that figure as reported rather than confirmed until xAI publishes its own benchmark numbers.
There are two ways a creator acts on this news today, and neither is "use Grok Build to make content" — because it cannot. The first is editorial: the agentic-coding race is exactly the kind of timely, high-intent story your audience is following right now. Drop your take on Grok Build 0.1 — how it stacks up against Claude Code and Codex, what $1/$2 pricing means for indie builders — into Kompozy as a source, and it fans that single point of view into a blog explainer, a comparison carousel, a few captioned short clips, and platform-native posts for X and LinkedIn, then schedules and publishes them across your channels. Being early and clear on a launch like this is how one opinion becomes a week of content.
The second is structural, for the builders in your audience: Grok Build ships software, Kompozy ships the marketing for it. If you use the model to build a product or a webhook-driven automation, Kompozy is the layer that turns each release into launch shorts, feature carousels, and a newsletter in your brand voice — and because it accepts a Custom Webhook as a publishing destination, the same agent that wrote your code can wire your changelog straight into your content pipeline. Grok Build handles the engineering; Kompozy handles the audience.
Grok Build 0.1 is xAI's coding model trained for agentic software engineering — writing, editing, and debugging code across a project. It is the model behind the Grok Build CLI and is now available to developers via the xAI API under the slug grok-build-0.1.
It is priced at $1.00 per million input tokens and $2.00 per million output tokens, with cached input at $0.20 per million — an 80% discount. It has a 256k-token context window and accepts text and image input.
No. It is a coding model and generates no images, video, or posts. To turn what you build into published content, pair it with a content engine like Kompozy that generates the media and publishes across nine platforms plus email and blog.
All three are terminal-native agentic coding agents. Grok Build is positioned as fast and lower-cost, with plan-first execution, parallel sub-agents in Git worktrees, and native MCP support. The right pick depends on your stack; treat head-to-head benchmarks as fast-moving snapshots.