Compute Grokbot — inference credits (POC)

Prepaid inference credits — Qwen2.5-14B via OpenAI-compatible API

Buy a $10 credit pack. Call POST /v1/chat/completions with your Bearer API key. Metered by tokens. Limited POC — first buyers only.

Buy $10 credits →

Checkout opens Polar (new tab). Button id=polar-checkout.

What you get

Item	Detail
Pack price	$10 USD one-time (Polar; platform fee ~10% → ~$9 usable credits)
Model	Qwen2.5-14B-Instruct (AWQ) behind our gateway
API	OpenAI-compatible `POST /v1/chat/completions` + `GET /v1/models`
Auth	Bearer API key delivered after payment
Metering	$0.70 / M input tokens · $1.40 / M output tokens
Caps	`max_tokens` clamped (default ≤ 256); rate limits + kill-switch may apply
Scope	Best-effort POC — no SLA; no PII / no sensitive prompts

Cold start: first request after idle can take ~5–8 minutes while the GPU worker comes up. Later requests are much faster.

How it works

Pay $10 via Polar (card).
Receive your API key and base URL.
Call the API like OpenAI Chat Completions; usage debits your balance. HTTP 402 when credits are exhausted.

curl -s "$BASE_URL/v1/chat/completions" \
  -H "Authorization: Bearer sk_poc_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"Qwen2.5-14B-Instruct-AWQ","messages":[{"role":"user","content":"hi"}],"max_tokens":64}'

What this is not

Not a RunPod account, GPU rental, or raw SSH/GPU access
Not a production SLA / multi-region cloud
Not for personal data, secrets, or illegal content

Terms

By purchasing you agree to our Terms of use (POC).

Seller: Compute Grokbot (temporary project name). Buyers must be 18+.