xAI's announcement and documentation confirm public API availability for Grok 4.7 and specify context, pricing and API behavior.
xAI releases Grok 4.7 through its public API
2026-09-21Verified announcementTRIAL
xAI released Grok 4.7 through its public API, Cursor, Grok Build and model routers. The model accepts text and images, exposes a 500K-token context window and supports tool use. Public API pricing starts at $2 per million input tokens and $6 output below 200K prompt tokens, while longer prompts double those rates.
Availability: public xAI API, Cursor, Grok Build and model routers
What this means for you
For teams choosing a coding or agent model, Grok 4.7 is directly callable and its long-context price discontinuity can materially change workload cost. Vendor benchmarks and safety claims remain self-reported, so compare task success, tool reliability and total spend on a fixed workload before routing production traffic.
Evidence & related material
Technical details & scores
| Availability | public xAI API, Cursor, Grok Build and model routers |
|---|---|
| Modalities | text and image input; text output |
| Context Window | 500,000 tokens; documentation lists no separate text-output cap |
| Pricing Under 200k Prompt Per Million | input 2 cached_input 0.5 output 6 |
| Pricing Over 200k Prompt Per Million | input 4 cached_input 1 output 12 |
| API Notes | US regional endpoint adds 10%; Responses API returns reasoning.encrypted_content |
| Fast Variant | same model at higher rate in Cursor and Grok Build; not offered through the public API |
| Time Basis | official announcement dated 2026-09-21; recovered as a late discovery from the fixed xAI monitor |
Technical impact 4
Engineering value 4
Adoption signal 3
Maturity 4
Verification cost 3
Risk 4