ByteDanceByteDance·🎬 Video Generation

Seedance 2.5 Edit

anonymized
Try on Venice.ai ↗
Quick reference
Seedance 2.5 Edit — TLDR
  • 🏢 ByteDance Seed's flagship video family, released 2026 as Seedance 2.5.
  • 🔧 Edit endpoint: handles audio and visual editing requests.
  • 📏 Up to 30-second audio-video clips in one generation.
  • 🆕 Clips can be extended twice for longer stories.
  • 👁️ Reference videos plus white-model (textureless 3D) guidance.
  • 🎬 Green-screen editing aimed at professional workflows.
  • 💬 Joint audio and picture generation in a single model.
  • 🔒 Proprietary model, available through API access in this catalog.
💰 Pricing
—
📅 On Venice since
Aug 7, 2026
87 days ago
Provider

ByteDance is a Chinese internet technology company headquartered in Beijing's Haidian district, with an associated variable-interest entity incorporated in the Cayman Islands. Best known as a consumer internet operator at enormous scale, it has also become a…

Read full profile →
28 models on Venice
22 video · 3 image · 3 inpaint
Since Nov 5, 2025

About this model

Seedance 2.5 Edit is the editing entry point into ByteDance's Seedance 2.5 video family, sitting alongside the text-to-video Seedance 2.5 T2V, image-to-video Seedance 2.5 I2V, and Seedance 2.5 R2V endpoints. It is the endpoint for the family's editing capabilities rather than for generating a clip from a prompt alone.

ByteDance Seed describes Seedance 2.5 as a next-generation audio-video joint generation model built for 30-second storytelling with precise reference control and editing capabilities. It follows the earlier Seedance 2.0 generation (and its lighter Seedance 2.0 Fast and Seedance 2.0 Mini tiers); the company reports single-pass generation of up to 30 seconds with the option to extend a clip twice, smoother and more consistent motion, and more realistic visuals.

For editing specifically, ByteDance says editing is more reliable and responds to a wider range of audio and visual editing requests. Reference handling is broadened too: reference videos are read for intention, framing and cinematic language rather than plain motion transfer, while white-model (textureless 3D) control and green-screen editing target professional production workflows.

Per this catalog's own listing, the model is proprietary and reached via API rather than self-hosted or fine-tuned, and it shares its 2026 release date with the other Seedance 2.5 endpoints. Earlier family members such as Seedance 1.5 Pro remain listed for comparison.

This About section is AI-generated from public sources (Claude Opus 5), with no human editing. It may contain inaccuracies — verify critical details against the sources listed above.

Data sources: Venice API · HuggingFace · Wikipedia — enrichment updated 15h ago