Cheap Grok API — xAI models from $0.65/1M
xAI's Grok line with reasoning and live web search — up to 68% below xAI's official price. Through /v1/responses the search tool returns real url_citation annotations — deep links to actual articles, not invented URLs — and search costs you input tokens instead of a per-query fee. Pay in ₽, USDT or by card, no VPN.
Grok API pricing
USD per 1M tokens, next to xAI's official rate. Updated automatically from our live price list.
| Model | Input | Output | xAI | Savings |
|---|---|---|---|---|
| Grok 4.5 | $0.65 | $2.00 | $2.00 | −68% |
| Grok 4.6 | $0.65 | $2.00 | $2.00 | −68% |
| Grok 4.7 | $0.65 | $2.00 | $2.00 | −68% |
| Grok Imagine Image | $0.02 / req | — | — | — |
| Grok Imagine Image Quality | $0.07 / req | — | — | — |
Why take Grok from here
Search with verifiable sources
Call /v1/responses with the web_search tool and you get url_citation annotations pointing at real, resolvable articles. Note that /v1/chat/completions drops the citation array — that's an endpoint limitation, so use responses when sources matter.
Search costs tokens, not a per-query fee
Search results are billed as input tokens like any other context. No separate charge per search the way first-party grounded APIs do it.
No model swapping
response.model always equals what you asked for. We don't quietly route a request for the flagship to a cheaper tier — check the field yourself on any call.
How to connect Grok
Any OpenAI SDK
Works in Cursor, Cline, Kilo Code, LangChain and anything else that lets you set a base URL.
from openai import OpenAI
client = OpenAI(
api_key="sk-...", # from cheapai.io/keys
base_url="https://cheapai.io/v1",
)Search with citations
Citations come back as annotations of type url_citation inside output[].content[].
curl https://cheapai.io/v1/responses \
-H "Authorization: Bearer sk-..." \
-H "Content-Type: application/json" \
-d '{"model":"grok-4-5",
"input":"what shipped in AI this week?",
"tools":[{"type":"web_search"}]}'Also available without a CDN: https://api.cheapai.io/v1 — use it for long requests where the model stays silent over ~2 minutes, or where the CDN is unreachable.
Full guides for every tool: documentation →
FAQ
How much does the Grok API cost here?
From $0.65 per 1M input tokens. 5 Grok models are live right now and the table above is pulled from the same price list the billing uses, so it is never stale.
Is this the real Grok, or a cheaper model behind the name?
The real thing. response.model always matches the model you requested, so you can verify it on any call without taking our word for it.
Do I get real source links from search?
Yes, via /v1/responses with the web_search tool: the answer carries url_citation annotations with deep links that actually resolve. /v1/chat/completions strips the citation array even when the model did search, so use responses when you need sources.
Do I need a VPN?
No. The domain and the API are reachable worldwide, and you can top up in ₽, USDT or by card.
Is there a subscription?
Not required. The default is pay-as-you-go from a balance that doesn't expire; monthly plans exist separately and give up to 40% off, but you never have to take one.