Grok Build 0.1
Grok Build 0.1 is xAI's coding model purpose-built for agentic software engineering workflows. It supports tool calling, structured outputs, text and image input, and a context window of 256K tokens, with always-on reasoning for multi-step development tasks.
import { streamText } from 'ai'
const result = streamText({ model: 'xai/grok-build-0.1', prompt: 'Why is the sky blue?'})Playground
Try out Grok Build 0.1 by xAI. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by xAI
| Model |
|---|
| 1M | 1.1s | 149tps | $1.25/M | $2.50/M | Read: $0.2/M Write: — | $5/K + input costs | — | +4 |   | 04/30/2026 | |||
| 2M | 0.4s | 128tps | $1.25/M | $2.50/M | Read: $0.2/M Write: — | $5/K + input costs | — | +3 | | 03/09/2026 | |||
| 2M | 1.8s | 1169tps | $1.25/M | $2.50/M | Read: $0.2/M Write: — | $5/K + input costs | — | +4 | | 03/09/2026 | |||
| 2M | 0.6s | 267tps | $1.25/M | $2.50/M | Read: $0.2/M Write: — | $5/K + input costs | — | +4 | | 03/09/2026 | |||
| 1M | 5.0s | 201tps | $0.20/M | $0.50/M | Read:$0.05/M Write:— | — | — | +3 | | 07/09/2025 | |||
| 1M | 0.4s | 86tps | $0.20/M | $0.50/M | Read:$0.05/M Write:— | — | — | +2 | | 07/09/2025 |
About Grok Build 0.1
Grok Build 0.1 was released May 20, 2026 as xAI's coding model purpose-built for agentic software engineering. Where general-purpose Grok models handle coding alongside other tasks, Grok Build 0.1 is trained to act as an autonomous engineering agent: refactoring code, invoking tools, producing structured outputs, and iterating through multi-step tasks until completion.
Reasoning is always active on Grok Build 0.1. There is no non-reasoning mode and reasoning effort is not configurable, so every response includes structured analysis before the final output. The model accepts text and image input, which means it can read diagrams, UI mockups, and error screenshots alongside source code. It supports tool calling and structured outputs, which fits agent loops that orchestrate file edits, test runs, and external APIs.
Grok Build 0.1 operates within a context window of 256K tokens and is available through Vercel AI Gateway at $1 per million input tokens, $2 per million output tokens, and $0.2 per million cached input tokens. The cached input rate makes it economical to reuse long codebase prefixes across many agent steps.
What To Consider When Choosing a Provider
- Configuration: Reasoning runs on every request. Grok Build 0.1 always produces chain-of-thought traces, which increases output token consumption compared to direct-response models. Budget output tokens generously and factor reasoning overhead into cost estimates.
- Configuration: The context window of 256K tokens fits substantial codebases in a single session, which suits agent loops that read many files before acting. Cached input pricing reduces the cost of reusing a long repository prefix across steps.
- Zero Data Retention: AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use Grok Build 0.1
Best For
- Autonomous Coding Agents: Loops that read files, call tools, and edit code without step-by-step prompting
- Codebase Refactoring: Multi-file changes that require reasoning over project structure within the context of 256K tokens
- Test Generation And Repair: Producing test cases, running them through tool calls, and patching failures
- Tool-Heavy Development Workflows: Pipelines that combine model decisions with shell commands, file edits, and external APIs
- Multimodal Engineering Tasks: Reading UI mockups, architecture diagrams, or error screenshots alongside source code
Consider Alternatives When
- Latency-Sensitive Completions: Always-on reasoning adds tokens and time per call. Grok 3 Fast or Grok 4 Fast Non-Reasoning return direct answers more quickly
- General-Purpose Text Tasks: Summarization, content generation, and conversation work better on Grok 3 Fast or Grok 4 Fast
- Image Or Video Generation: Use the Grok Imagine models for media output
- Maximum Reasoning Depth On Non-Coding Problems: Grok 4 and Grok 4.3 are tuned for math, science, and analytical reasoning beyond software engineering
Conclusion
Grok Build 0.1 is xAI's purpose-built model for agentic coding. Always-on reasoning, tool calling, structured outputs, and a context window of 256K tokens make it suitable for autonomous engineering loops rather than single-shot completions. Teams building coding agents should evaluate Grok Build 0.1 against Grok Code Fast 1 and the Grok 4 Fast variants for the right balance of agentic depth, latency, and cost.
Frequently Asked Questions
What is Grok Build 0.1 optimized for?
Grok Build 0.1 is purpose-built for agentic coding workflows. It's trained to act as an autonomous engineering agent that refactors code, invokes tools, produces structured outputs, and iterates through multi-step development tasks.
Does Grok Build 0.1 support a non-reasoning mode?
No. Reasoning is always active on Grok Build 0.1 and reasoning effort is not configurable. Every response includes structured analysis before the final output.
What inputs does Grok Build 0.1 accept?
Grok Build 0.1 accepts text and image input. You can include source code, diagrams, UI mockups, or error screenshots in the same request. Output is text.
What is the context window for Grok Build 0.1?
The context window is 256K tokens, which fits substantial codebases or long agent histories in a single session.
How does Grok Build 0.1 compare to Grok Code Fast 1?
Grok Build 0.1 is the next-generation coding model from xAI, with always-on reasoning, tool calling, and structured outputs designed for autonomous agents. Grok Code Fast 1 targets fast code generation and completion in interactive workflows.
What does Grok Build 0.1 cost through Vercel AI Gateway?
Current pricing is shown on this page. AI Gateway routes across providers, and rates may vary by provider.
How do I authenticate with Grok Build 0.1 through Vercel AI Gateway?
Use your Vercel AI Gateway API key with
xai/grok-build-0.1as the model identifier. AI Gateway handles provider routing automatically through the AI SDK plus Chat Completions, Responses, and Messages APIs.Does Vercel AI Gateway support Zero Data Retention for Grok Build 0.1?
Zero Data Retention is not currently available for this model. Zero Data Retention is offered on a per-provider basis. See https://vercel.com/docs/ai-gateway/capabilities/zdr for details.