ByteDance Seedance Video Model Family Comparison: How to Choose Across Eight Core Versions

Introduction: The Seedance Video Generation Matrix

ByteDance's Seedance family forms a layered video stack—from lightweight Lite models through professional Pro variants to the latest multimodal Seedance-2 series—covering basic generation and complex multi-reference workflows.

Eight routes often confuse teams: v1-lite-text-to-video, v1-lite-image-to-video, v1-pro-text-to-video, v1-pro-image-to-video, v1-pro-fast-image-to-video, seedance-1.5-pro, seedance-2-fast, and seedance-2. This guide maps each variant using complete API-shaped parameters so you can pick the right route fast.

Seedance tool hub: /home/seedance

I. Family Snapshot: One Table to Understand All Eight

No.	Model version	Core function	Input type	Duration	Standout	Positioning
1	v1-lite-text-to-video	Text to video	Prompt	5-10s	Up to 10k prompt chars	Entry
2	v1-lite-image-to-video	Image to video	1 image + prompt	5-10s	Optional end frame	Entry
3	v1-pro-text-to-video	Text to video	Prompt	5-10s	21:9 ultra-wide	Pro
4	v1-pro-image-to-video	Image to video	1 image + prompt	5-10s	Pro image quality	Pro
5	v1-pro-fast-image-to-video	Fast image to video	1 image + prompt	5-10s	Speed-first; 720p/1080p only	Fast Pro
6	seedance-1.5-pro	Text / image to video	0-2 images + prompt	4/8/12s	Audio gen, lens lock	All-round Pro
7	seedance-2-fast	Multimodal generation	Images/video/audio + prompt	4-15s	9 refs / 3 videos / 3 audios	Fast flagship
8	seedance-2	Multimodal generation	Images/video/audio + prompt	4-15s	1080p + web search	Top tier

II. Eight Models Deep Dive

Model 1: v1-lite-text-to-video

Entry-level text-to-video with minimal parameters and fast onboarding.

{
  "model": "bytedance/v1-lite-text-to-video",
  "prompt": "Text prompt (max 10000 chars)",
  "aspect_ratio": "16:9 / 4:3 / 1:1 / 3:4 / 9:16 / 9:21",
  "resolution": "480p / 720p / 1080p",
  "duration": "5 / 10",
  "camera_fixed": false,
  "seed": -1
}

Highlight: rare 9:21 tall vertical ratio for full-screen mobile framing.

Use cases: learning, rapid concept checks.

Route: /home/seedance/v1-lite-text-to-video

Model 2: v1-lite-image-to-video

Lite image-to-video with optional last-frame constraint.

{
  "model": "bytedance/v1-lite-image-to-video",
  "prompt": "Text prompt (max 10000 chars)",
  "image_url": "First-frame image URL",
  "resolution": "480p / 720p / 1080p",
  "duration": "5 / 10",
  "camera_fixed": false,
  "end_image_url": "Optional last-frame image URL",
  "seed": -1
}

Unique: end_image_url enables first/last frame guided motion.

Use cases: keyframe animation, animating stills.

Route: /home/seedance/v1-lite-image-to-video

Model 3: v1-pro-text-to-video

Pro text-to-video upgrade with cinematic ultra-wide support.

{
  "model": "bytedance/v1-pro-text-to-video",
  "prompt": "Text prompt (max 10000 chars)",
  "aspect_ratio": "21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16",
  "resolution": "480p / 720p / 1080p",
  "duration": "5 / 10",
  "camera_fixed": false,
  "seed": -1
}

Use cases: cinematic widescreen, higher-end text-only generation.

Route: /home/seedance/v1-pro-text-to-video

Model 4: v1-pro-image-to-video

Pro-grade image-to-video with stronger visual fidelity than Lite.

{
  "model": "bytedance/v1-pro-image-to-video",
  "prompt": "Text prompt (max 10000 chars)",
  "image_url": "First-frame image URL",
  "resolution": "480p / 720p / 1080p",
  "duration": "5 / 10",
  "camera_fixed": false,
  "seed": -1
}

Use cases: commercial-quality image-driven clips.

Route: /home/seedance/v1-pro-image-to-video

Model 5: v1-pro-fast-image-to-video

Speed-optimized Pro image-to-video; no 480p tier.

{
  "model": "bytedance/v1-pro-fast-image-to-video",
  "prompt": "Text prompt (max 10000 chars)",
  "image_url": "First-frame image URL",
  "resolution": "720p / 1080p",
  "duration": "5 / 10"
}

Use cases: batch jobs where turnaround matters.

Route: /home/seedance/v1-pro-fast-image-to-video

Model 6: seedance-1.5-pro

Unified Pro route: zero images behaves like text-to-video; one to two images behave like image-to-video.

{
  "model": "bytedance/seedance-1.5-pro",
  "prompt": "Text prompt (max 2500 chars)",
  "input_urls": ["Optional 0-2 image URLs"],
  "aspect_ratio": "1:1 / 4:3 / 3:4 / 16:9 / 9:16 / 21:9",
  "resolution": "480p / 720p / 1080p",
  "duration": "4 / 8 / 12",
  "fixed_lens": false,
  "generate_audio": false
}

Breakthroughs: dual-mode input; fixed_lens locks camera when needed; generate_audio adds native audio (often higher cost); durations 4/8/12 seconds.

Use cases: flexible creative pipelines without switching tools.

Route: /home/seedance/v1-5-pro

Model 7: seedance-2-fast

Fast multimodal flagship: rich references with speed bias and 480p/720p output.

{
  "model": "bytedance/seedance-2-fast",
  "prompt": "Optional prompt (max 20000 chars)",
  "first_frame_url": "First-frame image URL",
  "last_frame_url": "Last-frame image URL",
  "reference_image_urls": ["Up to 9 reference images"],
  "reference_video_urls": ["Up to 3 reference videos"],
  "reference_audio_urls": ["Up to 3 reference audios"],
  "generate_audio": true,
  "resolution": "480p / 720p",
  "aspect_ratio": "1:1 / 4:3 / 3:4 / 16:9 / 9:16 / 21:9 / adaptive",
  "duration": 5,
  "web_search": false
}

Reference rules: reference images plus first/last frames ≤ 9 total; ≤ 3 videos, combined duration ≤ 15s, each ≤ 50MB; ≤ 3 audio clips, combined ≤ 15s, each ≤ 15MB.

Use cases: character-consistent multimodal generation at speed.

Route: /home/seedance/v2-fast

Model 8: seedance-2

Full multimodal flagship: adds 1080p and web search on top of the Seedance-2 stack.

{
  "model": "bytedance/seedance-2",
  "prompt": "Optional prompt (max 20000 chars)",
  "first_frame_url": "First-frame image URL",
  "last_frame_url": "Last-frame image URL",
  "reference_image_urls": ["Up to 9 reference images"],
  "reference_video_urls": ["Up to 3 reference videos"],
  "reference_audio_urls": ["Up to 3 reference audios"],
  "generate_audio": true,
  "resolution": "480p / 720p / 1080p",
  "aspect_ratio": "1:1 / 4:3 / 3:4 / 16:9 / 9:16 / 21:9 / adaptive",
  "duration": 5,
  "web_search": false
}

Use cases: maximum quality multimodal briefs and research-grounded generation.

Route: /home/seedance/v2

III. Eight-Model Comparison Summary

3.1 Core feature matrix

Feature	Lite T2V	Lite I2V	Pro T2V	Pro I2V	Pro Fast	1.5-Pro	2-Fast	2
Text to video	✅	❌	✅	❌	❌	✅	✅	✅
Image to video	❌	✅	❌	✅	✅	✅	✅	✅
First/last frame	❌	✅	❌	❌	❌	❌	✅	✅
Multi-image refs	❌	❌	❌	❌	❌	❌	✅	✅
Video refs	❌	❌	❌	❌	❌	❌	✅	✅
Audio refs	❌	❌	❌	❌	❌	❌	✅	✅
Audio generation	❌	❌	❌	❌	❌	✅	✅	✅
Camera control	✅	✅	✅	✅	✅	✅	❌	❌
Web search	❌	❌	❌	❌	❌	❌	✅	✅
21:9 ratio	❌	❌	✅	❌	❌	✅	✅	✅
Max resolution	1080p	1080p	1080p	1080p	1080p	1080p	720p	1080p

3.2 Parameter complexity

Model	Required	Optional	Learning curve	Best for
Lite T2V	2	4	Easy	Beginners
Lite I2V	3	4	Easy	Beginners
Pro T2V	2	4	Easy	General users
Pro I2V	3	3	Easy	General users
Pro Fast I2V	3	2	Easy	Speed-first batches
1.5-Pro	2	6	Medium	Advanced creators
2-Fast	1	11	Complex	Power users
Seedance-2	1	12	Complex	Professional delivery

3.3 Duration support

Family	Allowed seconds	Range
Lite	5, 10	5-10
Pro (incl. Fast I2V)	5, 10	5-10
1.5-Pro	4, 8, 12	4-12
2-Fast / Seedance-2	4-15 (integer)	4-15

IV. Model Selection Decision Tree

What is your task?
|
|-- Generate from text only
|   |-- Quick onboarding -> Lite T2V
|   |-- Need 21:9 ultra-wide -> Pro T2V
|   |-- Need audio generation + lens lock -> 1.5-Pro
|   `-- Need top quality + web search -> Seedance-2
|
|-- Generate from one image
|   |-- Quality-first -> Pro I2V
|   |-- Speed-first -> Pro Fast I2V
|   |-- Need explicit first/last frames -> Lite I2V
|   `-- Need audio + lens + flexible duration -> 1.5-Pro
|
`-- Need multimodal references (images + videos + audio)
    |-- Speed priority -> Seedance-2-Fast
    `-- Quality priority (1080p) -> Seedance-2

V. Differentiators for Seedance

5.1 Full ladder from Lite to flagship

Seedance spans entry Lite models, professional Pro variants, unified 1.5-Pro, and multimodal Seedance-2—rather than forcing every workflow through one SKU.

5.2 First/last frame storytelling

Lite image-to-video plus Seedance-2 routes support explicit frame endpoints so motion can interpolate between defined poses.

5.3 Rich multimodal references

Seedance-2 family supports up to nine reference images combined with first/last frames, three reference videos, and three reference audio tracks—ideal for consistent characters and complex scenes.

5.4 Camera discipline

Lite, Pro, and 1.5-Pro expose camera or lens lock toggles for deliberate cinematography.

5.5 Optional web grounding

Seedance-2 routes expose web_search so prompts can lean on fresher real-world context when enabled.

VI. Final Recommendations

Scenario	Pick	Why
Fast entry text-to-video	Lite T2V	Minimal parameters, fast learning curve
Entry image-to-video with end frame	Lite I2V	Built-in last-frame constraint
High-quality text + 21:9	Pro T2V	Cinematic ultra-wide control
High-quality single-image video	Pro I2V	Pro-grade fidelity
Batch speed on image prompts	Pro Fast I2V	Optimized turnaround, HD-only
Unified text/image + audio + lens	1.5-Pro	Dual-mode input with advanced controls
Multimodal references (speed)	Seedance-2-Fast	Maximum reference breadth at 480p/720p
Multimodal references (quality)	Seedance-2	1080p plus web search when needed

One-line playbook:

Start with Lite when learning or validating ideas quickly.
Move to Pro when widescreen or commercial fidelity matters.
Use 1.5-Pro when one route must flip between text-only and light reference inputs with audio and lens control.
Choose Seedance-2 Fast vs Seedance-2 based on speed versus 1080p and web-assisted prompts.

All eight parameter stacks map directly to FuseAI Tools Seedance pages. Open the hub to browse every mode: /home/seedance.