Seedance 2.0
ByteDance video model with text-to-video, image-to-video, and video-reference generation, plus native audio, end-frame support, and flexible 4s to 15s timing.

Repair old, faded, or damaged images with a one-purpose restoration flow.

Turn portraits and product shots into clean graphite-style illustrations.

Turn a casual selfie into a polished business headshot for resumes, profiles, and team pages.

Clean up a product photo into a marketplace-ready packshot with a simple one-purpose flow.

Place a product into a polished campaign-style scene for ads, PDPs, and social creative.

Turn a clothing or accessory image into an editorial-style model photo for campaigns and lookbooks.

Reimagine a room photo into a designer-style interior concept without exposing complex controls.

Stage an empty or plain property photo with tasteful furniture and decor for listings and marketing.

Turn a product image into a polished paid-social creative built for scroll-stopping campaigns.

Create premium skincare and cosmetics campaign visuals from simple product photos.

Turn a jewelry image into a luxury campaign-ready visual for ecommerce and paid media.

Upgrade a dish photo into an appetizing restaurant, menu, or delivery-platform hero image.

Extract the print from a product photo and spin out a fresh pattern direction without rebuilding the setup by hand.

A flexible multi-model image workspace for prompts, references, and quick iteration.

OpenAI image workspace entry routed through Replicate, with prompt-based generation, reference images, quality/background controls, and up to 10 outputs.

Flagship image model for polished outputs, stronger detail, and fast concept work.

Balanced image model for everyday generation with a lower-cost default setup.

Replicate image model for prompt-based and reference-guided generation with 0.5 MP to 4 MP resolution tiers, custom width and height, and export-quality controls.

Fast image generation when speed matters more than exhaustive prompt tuning.

ByteDance image model for high-resolution prompt-based and reference-guided generation with 2K and 4K output options.

ByteDance image model for high-resolution prompt-based and reference-guided generation with a simple 4-credit cost.
Faster ByteDance video model for text-to-video and image-to-video generation with 2s to 12s timing.
ByteDance video model with text-to-video and image-to-video generation plus flexible 2s to 12s timing.
ByteDance video model with text-to-video, image-to-video, and video-reference generation, plus native audio, end-frame support, and flexible 4s to 15s timing.
xAI video model for text, image, and video-driven generation with flexible 1s to 15s timing and lightweight per-second pricing.
Image-to-video Kling model with standard or pro mode, 5s or 10s duration, and optional end frame control.
Flexible Kling video model with 3s to 15s generation, standard or pro mode, optional audio, and start or end frame control.
Premium video model with 4s, 6s, and 8s generations, optional audio, and support for multiple reference images.

Video model with flexible aspect ratios and 5s or 10s generation presets.