Grok: 4.7 Update, API Pricing and Buying Checks
Start with the product you need: a chat interface, a coding integration or API access. This update separates the September 2026 model announcement from older subscription information.
Visit GrokOctober 2026 update: Grok 4.7 on Gemini Enterprise Agent Platform
Preview — not general availability. Grok 4.7 is an xAI model. It is not a Google model. Google provides managed partner-model access through Gemini Enterprise Agent Platform / Model Garden. Google Cloud xAI access documentation.
The Grok 4.7 model card specifies model ID grok-4.7, text and image inputs, text output and a 524,288-token context. Function calling, structured output and reasoning are Preview capabilities. Fixed quota is supported; standard pay-as-you-go and Provisioned Throughput are not supported for this model on this platform.
Google Cloud docs checked 2026-10-03. The Preview contract, quotas and availability may change. Check your project's access and the model card before integration. The xAI API launch prices below do not establish Google Cloud platform pricing.
Grok 4.7 versus 4.6
The 21 September announcement says 4.7 is served at the same price and speed as 4.6, with changes aimed at longer coding and knowledge-work tasks. We have not independently measured those claims.
Access and capabilities
The announcement lists Grok API, Cursor and Grok Build among the access routes. Choose the product and model explicitly; availability in one integration does not prove identical limits in another.
API pricing
The launch starting rates are $2 per million input tokens and $6 per million output tokens. It also lists a fast variant at twice the price and output speed. Official announcement.
These are API launch figures, not a consumer subscription quote. Older X Premium and SuperGrok amounts have been removed because they were not verified as current.
Before adopting it
Compare a representative task set with explicit acceptance criteria. Check correctness, latency, token usage and any manual repair. Keep the same tools and starting context when evaluating a model change. A vendor benchmark is not a promise about your workflow.
Alternatives and related models
Compare the actual tasks you need rather than relying on a blanket model ranking. Browse AI chatbots, MiMo-V2.6 open checkpoints, or code review tools.
Frequently asked questions
What changed in Grok 4.7?
The official September 2026 announcement describes improved coding and knowledge-work capabilities relative to 4.6. Those are vendor claims, not results independently measured by AI Tool Finder.
How much does the Grok 4.7 API cost?
The launch announcement lists starting rates of $2 per million input tokens and $6 per million output tokens. A fast variant is listed at twice the price. Confirm the current API model and rates before purchase.
Is the API price the same as a chat subscription?
No. API usage and consumer subscriptions are different products. This page does not quote unverified current chat subscription amounts.
Should I switch from Grok 4.6?
Run the same representative tasks with the same acceptance criteria and compare correct outputs, latency and cost. The launch announcement alone does not establish the result for your workload.
Related guides
Browser automation tools · Browser agents versus API workflows
Built an AI Tool?
Submit your AI tool to be featured on AI Tool Finder and reach developers, founders, and productivity enthusiasts.
Submit Your AI ToolGrok chat and Grok Bot are separate
Grok Bot adds persistent cloud work, skills and routines. Its Shopify connection is installed inside Bot’s Connect apps and returns text; the Grok web connector does not install it there. Read the Bot profile for X access, inventory permissions and order-action exclusions. Shopify confirms the two product paths.