AI API Gateway + Video Generation

AI Video Generation + API Gateway + Traffic

Seedance 2.0 multimodal video generation, one API for 50+ AI models, and professional ad placement for enterprise customers.

Seedance 2.0 · 480p/720p · 24fps50+ AI Models · <50ms Latency · 99.9% SLA

Video Generation

Seedance 2.0 · Industry-leading quality

Multimodal video generation, editing, extension. Text / image / video / audio input, 24fps, 480p / 720p.

Use cases

Storyboards
From script to shot in seconds
Ad creatives
Product demos, TVC, social ads
Digital humans
Virtual presenters, corporate, training
Brand campaigns
Marketing, seasonal, IP animation

Three steps, pay as you go

No subscription, no minimum, failed calls not charged

1
Sign up
Email or phone, 30 seconds
2
Top up
Any amount, balance never expires
3
Call the API
OpenAI-compatible — just change base_url

AI Short-Drama Production

Reel · From one idea to a full production package

Script, characters, scenes, storyboard — end-to-end AI direction. No editing experience needed. Free to use.

A four-step pipeline

01
Script
Generate episode outlines and scene-by-scene dialogue from a single line
02
Characters
Visual design, costume variants, and prompts — consistent across episodes
03
Scenes
Auto-generated environments, lighting and mood matched to the script
04
Storyboard
Shot list, camera notes and ready-to-render Seedance prompts
Try Reel — Free

No install, ready in seconds

Platform

Built different.

Not another proxy. A full execution platform with intelligent routing, edge compute, and pricing that passes every cost reduction to you.

INFRASTRUCTURE

Global Edge Execution

Every API call executes on the nearest edge node. 300+ points of presence worldwide. Sub-50ms routing latency. Your requests never traverse unnecessary hops.

Powered by Cloudflare Workers. Automatic failover between regions. Auth caching at the edge via KV. Billing logs written asynchronously.

300+

Edge Nodes

<50ms

Latency

INTELLIGENCE

Auto Model Routing

Set model to "auto" and let the router pick the optimal model for each task. Simple queries route to cheap models. Complex reasoning routes to powerful ones.

40-60%cost savings

Reduces total cost by 40-60% vs. always using a premium model. Task classification runs at the edge with zero added latency. You keep full control with per-request overrides.

PRICING

Transparent Cost Advantage

Pay per use at rates consistently below official pricing. When upstream providers cut prices, your rates drop automatically. You always benefit from the AI cost curve.

Off-peak 0.7xNo expiry

Per-token billing granularity. Off-peak discount (0.7x during 00:00-08:00 UTC+8). Price increases require 7-day advance notice. No expiration on pay-as-you-go balances.

Seedance 2.0 Pricing

Estimate Seedance 2.0 video tokens

Adjust the model, resolution and duration to estimate tokens used per call. Actual usage matches what we deduct from your wallet on completion.

tokens = (input_duration + output_duration) × width × height × fps / 1024

Formula: (input_dur + output_dur) × width × height × fps / 1024. Actual usage matches usage.completion_tokens returned by the API.

Estimator

5s
3s15s
(0 + 5) × 1280 × 720 × 24 / 1024 = 108,000
Estimated tokens108,000
1280×720 · 720p · 5s · 24fps