Video Generation

Generate video from text or images through one API — Sora 2, Seedance, character animation and speaking video. Pay per clip in USDC via x402, no subscription.

Generate video from text or images through one API — Sora 2, Seedance, character animation and speaking video. Pay per clip in USDC via x402, no subscription.

Every endpoint below is callable with one API key or paid per call in USDC over x402. There is no subscription and no minimum spend; you pay for the calls you make.

Sora 2

OpenAI Sora 2 via Azure AI Foundry, 720p + synced audio

OpenAI Sora 2 via Azure AI Foundry. Generate 4/8/12-second video clips at 720p with synchronized audio from text prompts. Portrait or landscape. Async workflow: submit prompt → poll status → get MP4 URL. Works with Image Generation (use generated images as video seeds) and Music Generation (add soundtrack to clips).

Pricing: $0.421001 per clip

Seedance Video

ByteDance video generation, 720p + synced audio

ByteDance video generation at 720p with synced audio. Model tiers, each quoted separately on the same route: 1.5 Pro ($0.354916 — budget, image-to-video only), 2.0 Mini ($0.399554), 2.0 Fast ($0.826804 — 60-80s generation, supports RealFace), 2.0 ($1.13648 — supports RealFace), 2.5 ($1.578875 — longest clips, up to 30s). Two input modes: (1) Text-to-video: describe a scene in words; (2) Image-to-video: upload a photo/illustration as the first frame, the AI animates it. Character consistency: enroll via Virtual Portrait or RealFace → get ta_xxx asset_id → pass it to every generation for the same character across all clips.

Pricing: $0.354916–$1.578875 per clip

Grok Imagine Video

xAI video generation from text prompts

xAI Grok Imagine video generation. Two tiers, quoted separately on the same route: base ($0.402) and 1.5 ($0.642). Same async workflow as the other video models — submit a prompt, poll status, collect the MP4 URL.

Pricing: $0.402–$0.642 per clip

Seedance Virtual Portrait

AI character enrollment — keep same character across clips

Enroll a synthetic AI character to get a ta_xxx asset_id. Upload a front-facing character image (JPG/PNG/WEBP, under 10MB — illustration, 3D render, AI-generated face). No real person needed, no KYC, no liveness check. One-time $0.011 enrollment. Then pass the asset_id in the real_face_asset_id parameter to any Seedance 2.0 video generation — your character stays consistent across all clips. Ideal for brand mascots, virtual influencers, game characters, animated series.

Pricing: $0.011/enrollment (one-time)

Seedance RealFace

Real-person enrollment with on-phone liveness check

Enroll a real person for character-consistent video generation. Full flow: (1) Connect wallet; (2) Provide a direct link to a clear front-facing photo (JPG/PNG/WEBP, under 10MB); (3) The person scans a QR code on their phone, grants camera access, and performs a 2-4 second liveness check (nod + blink) — no government ID needed; (4) Pay $0.011 USDC; (5) Receive ta_xxx asset_id instantly. Biometric data never touches our servers — it goes directly to the verification partner and is discarded. The asset_id works with Seedance 2.0 Fast and 2.0 Pro only (not 1.5 Pro). Only the enrolled person can appear in generated videos. Suitable for personalized marketing, corporate training, content creators.

Pricing: $0.0055–$0.011 per step

How to pay

Call any endpoint. Without a key you get HTTP 402 with the exact price in USDC; sign that quote with a Base or Solana wallet and repeat the request. With an account API key the same call is billed to your balance instead.

curl -i https://api.jarvisclaw.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"auto/free","messages":[{"role":"user","content":"Hello"}]}'