kino402

When to call

• Need other via x402 and want Cleared-indexed payTo with a live scorecard.
• Budget between $0.001 and $0.309 per call on published endpoints.
• kino402 sells frontier text-to-video and text-to-speech per call, in USDC on Base or Solana over x402. No account, no API key, no minimum. THE QUOTE IS THE 402. POST an unpaid request to /v1/generate or /v1/audio/tts and the 402 that comes back carries the exact price for the body you sent — not a rate card, the number for that call. Pay it by repeating the identical request with the payment header, and the amount that settles is the amount that was quoted. Headers: send X-PAYMENT for x402 v1, or PAYMENT-SIGNATURE for x402 v2. The v1 envelope is the 402 response body; the v2 envelope is standard base64 in the PAYMENT-REQUIRED response header. Both versions are served from the same 402, so use whichever your client speaks. Rails: Base — USDC 0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913, network "base" in x402 v1 and "eip155:8453" in v2; Solana — USDC EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v, network "solana" in x402 v1 and "solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp" in v2. Every rail carries the SAME quote for a given request — a second rail is a second way to pay, never a second price. The accepts array in the 402 you were handed is authoritative; pay the first entry you can, which is the order they are listed in. POST is how you buy. A GET or HEAD on a paid route answers a reference 402 so a prober can read the terms without spending, but a payment presented on GET is refused without settling. /v1/generate is asynchronous. A paid POST returns {job_id, status_url} once the render is accepted upstream. Poll GET /v1/jobs/{id} — free and unauthenticated — until status is "completed" (the mp4 is at video_url) or "failed". A clip usually takes 1-6 minutes. The video_url expires: download it promptly. /v1/audio/tts is synchronous. A paid POST returns the mp3 in the response body. A /v1/forms/ route sells a FORMAT rather than a model: you send a structured brief — for the UGC form, a product name and one to five selling points — and the service compiles it into the prompt and renders it on your behalf. It is priced as the renderer plus a compile fee, quoted in the same 402, delivered as the same job id, and the compiled prompt comes back with it so you can inspect or reuse the shot. GET /v1/models publishes every field of every form with its bounds and its default. A /v1/prompts/ route sells that compiler on its own, flat-priced and synchronous: the same brief in, and {prompt, render_plan, compiler} straight back in the response body — no job, no polling, no video. It is the cheapest thing in the catalog and it is the honest way to evaluate the format before buying a render, or to buy the prompt craft alone if you already have renderer credits. The prompt it returns is a valid `prompt` for POST /v1/generate, and render_plan carries the model, seconds, resolution and aspect ratio it was written for. GET /v1/models carries the full rate card: every model id, its parameter ranges, the rate formula it is priced from, and a worked example price. Read it before you pick a model — the cheapest and the most expensive video call in the catalog differ by more than two orders of magnitude.

Risks

• Found — not operator-owned; claim status unknown.
• No Cleared settlement receipt on file yet.
• No Gateway traffic yet — market share unproven.