grok-4.7
xAI's frontier reasoning model, with optional real-time grounding in the X firehose for current events. Tuned for hard analytical problems and up-to-the-minute awareness.
At a glance
Released
Oct 1, 2026
Provider
xAI
Context window
256,000 tokens
Max output
64,000 tokens
Modalities
Text + image → text
Knowledge cutoff
September 2026
API model ID
grok-4.7License
Proprietary (API only)
Pricing
Standard
Input$2 / 1M
Output$6 / 1M
Long-context requests are billed at $4 in / $12 out per 1M. A typical 1,500-token prompt with a 650-token answer costs ≈ $0.007 on the standard tier. Run your own prompt through the calculator →
What's new vs the previous model
Compared withgrok-4.6 (August 2026)
- Sharper multi-step reasoning and more reliable tool use
- Real-time retrieval can add latency on grounded queries
Best use cases
Current-events analysis
Answers questions grounded in live X data and breaking news.
Hard reasoning
Works through math, logic and complex coding problems step by step.
Research agents
Drives agentic workflows that combine reasoning with real-time retrieval.
Strengths & limitations
Strengths
- Strong reasoning
- Real-time grounding via X
- Reliable tool calling
Limitations
- Image-only beyond text, no audio or video
- Grounded queries can be slower
How to test it
- No-code: open grok.com and pick grok-4.7 from the model switcher — no API key needed.
- Playground: sign in to the xAI Console and run prompts in the workbench, then copy the generated code.
- API: create a key in the xAI Console and call the API (example below).
curl https://api.x.ai/v1/chat/completions \
# OpenAI-compatible endpoint
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-4.7",
"messages": [{"role": "user", "content": "Explain tokenization in one paragraph."}]
}'