VideoAnonymized

Seedance 2.0 Mini R2V

ByteDance's lightweight reference-to-video model — supports text, image, video, and audio inputs for 15-second clips at 720p, optimized for speed and cost.

Maker
ByteDance
Modality
Video + audio
Max duration
15 seconds
Max resolution
720p

Overview

What is Seedance 2.0 Mini R2V

Seedance 2.0 Mini R2V is ByteDance's compact reference-to-video model, released in June 2026 as part of the Seedance 2.0 family. It generates up to 15-second video clips using text prompts combined with reference inputs like images, video clips, and audio, supporting native audio-video sync and aspect ratios from 21:9 to 9:16.

Running it privately on Venice

On Venice, Seedance 2.0 Mini R2V runs under an anonymized privacy tier — your prompts are not stored, profiled, or used for training. This enables permissionless, uncensored video generation while preserving sovereignty over your creative inputs. You pay per clip, with no subscription, making it ideal for testing ideas at scale without long-term commitment.

AnonymizedNo prompt trainingTEE · hardware enclaveEnd-to-end encrypted

Assessment

Strengths and limitations

Strengths
  • Cost-efficient for high-volume short-form video generation — ideal for rapid iteration and A/B testing.
  • Supports multimodal reference inputs: combine text, images, video clips, and audio to guide motion, composition, and sound.
  • Native audio-video generation with synchronized output, enhancing realism for social and commercial content.
  • Fast generation speed: roughly twice as fast as Seedance 2.0 Fast, enabling quick turnarounds.
  • Preserves camera motion and style from references, useful for consistent branding and cinematic continuity.
Limitations
  • Capped at 720p resolution: not suitable for high-fidelity 1080p or 4K deliverables.
  • Limited to 15-second clips, restricting use for longer narratives or explainers.
  • Proprietary and closed: no open weights, so self-hosting or fine-tuning is not possible.
  • Less detailed output compared to higher-tier Seedance models, especially in complex scenes.

Capabilities

What it supports

  • Image to video
  • Reference to video
  • Native audio generation

Specifications

Datasheet

Maker
ByteDance
Released
June 15, 2026
Modality
Reference-to-video (text, image, video, audio inputs)
Resolutions
720p, 480p
Clip lengths
4s – 15s
Mode
image-to-video
Aspect ratios
21:9, 16:9, 4:3, 1:1, 3:4, 9:16
Audio
Yes
Prompt limit
10,000 chars
Privacy on Venice
Anonymized — prompts not stored
Available on Venice since
Jun 2026
License
Proprietary

API

Call it from your code

Venice exposes this model through the REST API. Queue a generation with the model id.

curl https://api.venice.ai/api/v1/video/queue \
  -H "Authorization: Bearer $VENICE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-0-mini-reference-to-video-basic",
    "prompt": "Aerial drone shot over a misty mountain valley at golden hour"
  }'

# Use the returned queue_id with https://api.venice.ai/api/v1/video/retrieve.
# Call /video/complete after downloading if needed.

Pricing

What it costs on Venice

Pay per clip on Venice — price scales with resolution and duration (4s–15s), from $0.18.

720p · 4s
$0.38
Per clip
480p · 4s
$0.18
Per clip

New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.

Alternatives

How it compares

ModelMax resolutionStrongest atOpen weightsPrice (Venice)
Seedance 2.0 Mini R2V720pMultimodal reference inputNofrom $0.18
Wan 2.7Open-source video generationYesfrom $0.55
Kling O3 ProLonger, high-fidelity clipsNofrom $0.46
Vidu Q3Photorealistic detailNofrom $0.27

Best for fast, low-cost reference-guided clips with audio sync.

Use cases

What it is good for

  1. 01Short-form social content for TikTok, Instagram Reels, and YouTube Shorts.
  2. 02Rapid prototyping of ad concepts using reference materials and background audio.
  3. 03Product reveal videos with consistent character and scene continuity.
  4. 04Music-driven visualizers using reference audio and image-to-video transitions.
  5. 05Multi-shot cinematic sequences guided by reference frames and camera language.

Prompting

Getting better results

Use @Image1, @Image2 in your prompt to reference uploaded images by index.

Include camera direction cues like 'slow zoom' or 'wide-angle shot' for better motion control.

For audio sync, describe rhythm or mood: 'synced to upbeat tempo' or 'matching the drumbeat'.

Keep prompts under 10,000 characters — focus on key actions and transitions within 15 seconds.

Version history

Seedance 2.0
2026-04

Flagship tier with 1080p support and higher input limits.

Seedance 2.0 Fast
2026-05

Faster, lower-cost draft model.

Seedance 2.0 Mini R2V
2026-06

Current — compact, cost-efficient reference-to-video.

FAQ

Frequently asked questions

Seedance 2.0 Mini R2V is ByteDance's lightweight reference-to-video model, released in June 2026. It generates up to 15-second video clips using text prompts combined with reference images, videos, and audio, supporting native audio-video sync and multiple aspect ratios.

On Venice, pricing starts at $0.18 for a 480p, 4-second clip, scaling with resolution and duration up to $1.42 for longer 720p outputs. You pay per generation with no subscription required.

No. Seedance 2.0 Mini R2V is a proprietary model developed by ByteDance. It is not open source, and weights are not publicly available for self-hosting or modification.

It supports 720p and 480p resolutions, with aspect ratios including 21:9, 16:9, 4:3, 1:1, 3:4, and 9:16, making it flexible for cinematic, social, and mobile formats.

Yes. It natively generates synchronized audio-video output, allowing background music or sound effects to be included in the final clip based on reference audio or prompt cues.

Yes. It supports reference-to-video generation using text, images, video clips, and audio. You can upload reference materials and refer to them in the prompt using @Image1, @Image2, etc.

Seedance 2.0 Mini R2V excels in multimodal reference input and audio sync, ideal for commercial short-form content. Wan 2.7 is open-source and customizable but lacks native audio and reference-to-video capabilities, making it better for developers than creators.

Run Seedance 2.0 Mini R2V privately

No prompt logging. No data used for training.