Free Grok 4.6 via the Tokenator API

What the model is for

This model is available on a free plan with set usage limits. Grok 4.6 is SpaceXAI's most intelligent model, demonstrating cutting-edge results in programming, intellectual work, and STEM education.

Specification

Full nameFree Grok 4.6
API IDfree-grok-4.6
Model vendorxAI
TypeText model (chat / completions)
Context window500K tokens
Max output128K tokens
Image inputyes
Tokenator usage multiplierdoes not draw down the key limit4M tokens per day60 requests per dayresets at 00:00 UTC
StabilityUnstablethe model may need a retry or cut an answer short
Input modalitiestext, image, file
Aliasescursor-free-grok-4.6
Supported API formats/v1/chat/completions /v1/responses /v1/messages
Upstream providers2
Current statusavailable
Data updated2026-08-22

Upstream providers

ProviderMultiplier
Free Grok Partner #11.8×
LimitedAI1.8×

A request goes to the first available provider by priority; if it fails, Tokenator switches to the next one.

Example request

curl
curl https://api.tokenator.top/v1/chat/completions \
  -H "Authorization: Bearer sk-your-tokenator-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "free-grok-4.6",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Compatible tools

The model can be used in any tool that accepts a custom base URL: Claude Code, Codex CLI, OpenCode, Cursor, Cline, Kilo Code, Cherry Studio, OpenClaw.

FAQ about Free Grok 4.6

What is the API ID of Free Grok 4.6?

free-grok-4.6 — put this into the model field of your request.

What context window does Free Grok 4.6 have?

500K tokens; max output is 128K tokens.

How many tokens will a request cost?

(input + output) × 1.8×. See the documentation for details.

Which tools support this model?

Any tool that accepts a custom base URL: Claude Code, Codex CLI, OpenCode, Cursor, Cline, Kilo Code. Setup guides are in the integrations section.

Is Free Grok 4.6 really free?

Yes. Requests to this model do not draw down your key's token limit — no multiplier is applied and your paid bundle stays untouched. A daily allowance applies instead: 4M tokens and 60 requests per day.

What is the limit on the free model and when does it reset?

The allowance — 4M tokens and 60 requests per day — is counted per key and per model, and resets at 00:00 UTC. Spending one free model's limit leaves the other free models usable.

What happens when the daily limit runs out?

The request returns 429 with the code free_model_daily_limit and a message stating what was spent and when it resets. Your paid key limit is left alone — there is no automatic switch to paid billing. The remaining allowance per free model is returned by /v1/tokens.

Do free models need a separate key?

No. The same key and the same base URL as for paid models — only the model field changes. The full list of free models is on its own page, and the mechanics are in the documentation.