COMMUNITY PAGE
Run Seedance 2.0 Mini on Floyo
Home / Model / Seedance 2.0 Mini on Floyo
AI VIDEO GENERATION
Run Seedance 2.0 Mini on Floyo
ByteDance's high-volume video tier. Same Seedance 2.0 architecture at half the cost and 2x the speed. Cinematic multi-shot output with native audio, camera control, and character consistency. 720p at 4-15 seconds. Built for rapid iteration and bulk production.
Run ByteDance's Seedance 2.0 Mini through ComfyUI in your browser. No API key, no installs, no local GPU.
|
Speed 2x faster than Seedance Fast |
Cost 50% of standard Seedance 2.0 |
|
Resolution 480p / 720p |
Duration 4-15 seconds per clip |
No installation. Runs in browser. Updated July 2026.
What You Get
Seedance 2.0 Mini is ByteDance's lightweight video generation tier, released June 2026. Same Dual-Branch Diffusion Transformer architecture as standard Seedance 2.0 with reduced compute: 2x faster generation and 50% lower cost. Cinematic multi-shot output with camera control (pans, zooms, tracking shots), character consistency across scenes, and native audio synthesis. Accepts text prompts, reference images (up to 9), video clips, and audio. Outputs at 480p/720p in 16:9, 9:16, 1:1, 4:3, and 3:4 aspect ratios, 4-15 seconds per clip. Built for high-volume creation: social clips, ad variants, concept tests, and rapid iteration. Available as a ComfyUI API node on Floyo.
What is Seedance 2.0 Mini?
Seedance 2.0 Mini is the lightweight, high-volume tier of ByteDance's Seedance 2.0 video family. Released in June 2026, it uses the same unified Dual-Branch Diffusion Transformer architecture as standard Seedance 2.0 but with reduced compute per generation. The result: roughly 2x faster than Seedance 2.0 Fast at 50% of the standard tier's cost. It is a sibling tier, not a separate model family.
"Mini" refers to cost and speed, not to stripped capabilities. The model still accepts text prompts, single images as a first frame, and reference material (up to 9 reference images, plus reference videos and audio on supported platforms). Multi-shot direction, timeline prompting, camera control, and character consistency all carry over from standard Seedance 2.0. The inputs are intact. Mini is a cheaper way to do the same jobs.
The trade-off is resolution and polish. Mini outputs at 480p and 720p. Standard Seedance 2.0 goes up to 1080p and 2K. For social clips, TikTok content, Instagram Reels, ad concept tests, and rapid iteration where you need 50 variations by Friday, Mini is the correct tier. For a polished 15-second hero spot at 2K with perfect lip-sync, step up to standard Seedance 2.0.
Seedance 2.0 itself is the top-performing model on the Artificial Analysis Video Arena for both text-to-video and image-to-video as of early 2026. Mini inherits that quality baseline but trades maximum fidelity for throughput. The practical read: prototype with Mini, promote your winners to the standard tier for final delivery.
On Floyo, Seedance 2.0 Mini runs through ComfyUI API nodes on H100 NVL GPUs. Write a prompt, optionally upload reference images, and generate. No Dreamina account, no ByteDance API key, no Volcano Engine setup.
What are Seedance 2.0 Mini's technical specifications?
Seedance 2.0 Mini uses ByteDance's Dual-Branch Diffusion Transformer architecture (same as Seedance 2.0) with reduced compute. Outputs at 480p/720p in 5 aspect ratios, 4-15 seconds per clip. Accepts text, image, video, and audio references. Native audio synthesis for synchronized dialogue and sound effects. 2x faster than Seedance 2.0 Fast. 50% cost of standard Seedance 2.0 (~0.5 RMB per second at 720p, roughly $0.073/second).
| Spec | Details |
|---|---|
| Developer | ByteDance (Seed Research Team) |
| Architecture | Dual-Branch Diffusion Transformer (shared with Seedance 2.0, reduced compute) |
| Family | Seedance 2.0 Standard / Fast / Mini (Mini is the lightweight tier) |
| Resolution | 480p, 720p |
| Duration | 4-15 seconds per clip |
| Aspect Ratios | 16:9, 9:16, 1:1, 4:3, 3:4 |
| Inputs | Text prompts, reference images (up to 9), video clips, audio |
| Audio | Native synchronized audio (dialogue, SFX, ambient) |
| Multi-Shot | Yes (structured sequences with camera direction and scene transitions) |
| Camera Control | Pans, zooms, tracking shots, dynamic framing (prompt-driven) |
| Character Consistency | Yes (reference-guided identity preservation across shots) |
| Speed vs Standard | ~2x faster than Seedance 2.0 Fast, ~5x faster than full Seedance 2.0 |
| Cost vs Standard | 50% of standard Seedance 2.0 (~$0.073/second at 720p) |
| Arena Ranking (Parent) | Seedance 2.0: #1 T2V + I2V on Artificial Analysis (early 2026) |
| ComfyUI Access | API-based node on Floyo |
| Release Date | June 2026 |
What can you create with Seedance 2.0 Mini?
Seedance 2.0 Mini covers social media clips, ad concept tests, product videos, daily content, entertainment shorts, brand campaigns, educational explainers, travel and food clips, and any high-volume video production where cost-per-clip matters more than maximum resolution. Prototype with Mini, ship the winners, promote finalists to standard Seedance 2.0 for polish.
| Capability | What It Does | Use Case |
|---|---|---|
| Cinematic Multi-Shot | Structured sequences with pacing, shot direction, camera transitions, and narrative flow. Not isolated clips. Multi-shot storytelling in a single generation. | Short films, brand stories, series content, ad campaigns |
| Camera Motion Control | Prompt-driven pans, zooms, tracking shots, dynamic framing, and scene transitions. Describe the camera move and the model executes it. | Product reveals, architectural walkthroughs, cinematic openers |
| Character Consistency | Reference-guided identity preservation. Upload reference images and the same character holds across multiple shots, scenes, and compositions. | Brand mascots, character-led campaigns, recurring hosts |
| Native Audio | Synchronized dialogue, sound effects, and ambient audio generated in the same pass as video. No separate audio model or manual sync. | Talking head content, product demos, immersive scenes |
| High-Volume Iteration | Generate 50 variations for the price of 25 on standard Seedance 2.0. Test hooks, angles, styles, and formats quickly. Promote winners to the standard tier for final delivery. | A/B testing, campaign exploration, social content calendars |
| Pipeline Integration | Chain with other models in ComfyUI. Generate a character with Nano Banana, animate with Seedance Mini, add custom narration with ElevenLabs, upscale with Topaz. | Multi-model production pipelines |
How does Seedance 2.0 Mini compare to other video models?
Seedance 2.0 Mini leads on cost-per-clip for cinematic quality video. HappyHorse 1.0 leads on arena ranking and cinematic depth-of-field. Vidu Q3 leads on duration (16 seconds) and reference consistency. Wan 2.7 leads on open-source flexibility. Standard Seedance 2.0 leads on resolution (up to 2K) and final delivery quality. Mini's edge: the same Seedance 2.0 capabilities at half the price, built for the volume work that eats most production budgets.
| Model | Resolution | Native Audio | Cost Tier | Best For |
|---|---|---|---|---|
| Seedance 2.0 Mini | 480p/720p | Yes | 50% of standard ($0.073/s) | Volume + iteration |
| Seedance 2.0 Standard | Up to 2K | Yes | Full price | Final delivery + polish |
| HappyHorse 1.0 | 1080p | Yes (7-lang lip-sync) | API pricing | Cinematic DOF + arena #1 |
| Wan 2.7 | Up to 4K | No | Open source (free) | Open-source flexibility |
Source: ByteDance Seed official materials, Volcano Engine FORCE conference (June 23, 2026), Artificial Analysis Video Arena, WaveSpeedAI documentation, Atlas Cloud performance review, Pexo explainer, and SeedanceTips pricing guide as of July 2026.
What are Seedance 2.0 Mini's key features?
Seedance 2.0 Mini's feature set is identical to standard Seedance 2.0 in capability, narrower in resolution. The same camera control, the same character consistency, the same multi-shot direction, the same audio synthesis. What changes is speed and cost. This makes it the correct tool for the 80% of video production work where final 2K resolution is not the deliverable.
2x Speed, 50% Cost
The headline numbers. Seedance 2.0 Mini generates roughly 2x faster than Seedance 2.0 Fast (the previous speed tier) and costs about half of standard Seedance 2.0. At 720p, that works out to approximately $0.073 per second of video. For teams generating dozens or hundreds of clips per week, this changes the unit economics of AI video production from "expensive experiment" to "default workflow."
Cinematic Multi-Shot Sequencing
Seedance 2.0 Mini generates structured scenes with narrative flow, not isolated clips. Shot direction, camera transitions, pacing, and scene composition are all handled within a single generation. A 15-second output can feel like an edited sequence with establishment shots, close-ups, and resolution. This is inherited directly from the Seedance 2.0 architecture.
Multimodal Reference Input
Feed the model text prompts alongside up to 9 reference images, reference video clips, and audio references. These guide subject appearance, movement style, mood, pacing, and visual continuity. Existing brand assets, product photos, and footage clips can all steer the output rather than relying on text description alone.
Character Consistency Across Shots
Upload reference images and the same character identity holds across multiple shots and scenes. This is what makes Mini usable for brand content and series work. A mascot or product ambassador looks the same in every clip. Combined with multi-shot sequencing, you can build short narrative arcs with visual continuity that feels produced, not generated.
Native Audio Synthesis
Dialogue, sound effects, and ambient audio are generated alongside video in the same pass. No separate audio model, no manual synchronization. The Dual-Branch architecture produces audio and video simultaneously, which is why lip-sync and environmental sounds stay coherent. This is the same audio pipeline as standard Seedance 2.0.
Social-Native Format Support
All five standard aspect ratios (16:9, 9:16, 1:1, 4:3, 3:4) output as ready-to-post MP4. Vertical for TikTok and Reels, square for feeds, widescreen for YouTube. No post-export cropping or reformatting. Generate in the format you are publishing in.
How does Seedance 2.0 Mini work?
Seedance 2.0 Mini uses the same Dual-Branch Diffusion Transformer architecture as standard Seedance 2.0. The dual branches generate video and audio simultaneously in a single forward pass, which is why dialogue, sound effects, and ambient audio stay synchronized with the visual output. Mini trims the compute per generation to achieve faster throughput and lower cost while keeping the core capabilities intact.
The model processes your inputs (text prompt, reference images, reference video, audio) and generates a structured video sequence. Camera movements specified in the prompt (pans, zooms, tracking shots) are executed as generated motion, not post-processing effects. Multi-shot transitions happen within the generation itself, following the narrative structure implied by the prompt.
Character consistency uses reference-guided identity extraction. The model reads facial features, clothing, body proportions, and distinguishing visual traits from your reference images and maintains them across every frame and shot in the output. This is learned behavior from the Seedance 2.0 training pipeline, not a separate identity network.
On Floyo, Seedance 2.0 Mini runs through ComfyUI API nodes on H100 NVL GPUs. Your inputs are sent to the inference servers, and the generated video with audio returns as an MP4 to your ComfyUI canvas. You can chain it with other models: generate a character image with Nano Banana, animate with Seedance Mini, add custom narration with ElevenLabs or Fish Audio S2, upscale the final output with Topaz.
Fair warning: Seedance 2.0 Mini outputs at 480p/720p. It does not support 1080p or 2K. For final delivery at higher resolution, use standard Seedance 2.0 or upscale Mini output with Topaz Video AI on Floyo. Audio quality may not match the tightest lip-sync of the standard tier. Content filtering is active. Commercial use terms depend on the access route (Floyo, Dreamina, BytePlus all carry different clauses). API pricing applies through your Floyo API Wallet.
Frequently Asked Questions
Common questions about running Seedance 2.0 Mini on Floyo.
You can start with Floyo's free pricing plan. Floyo gives $0.25 in free API credits on signup. To continue using the service beyond the free tier, upgrade your Floyo pricing plan. Seedance 2.0 Mini runs as an API node, so generation costs come from your API Wallet (separate from your plan's GPU time). At roughly $0.073 per second of 720p video, it is the most cost-effective way to use the Seedance 2.0 architecture.
Open Floyo in your browser, search "Seedance" in the template library, and pick the Seedance 2.0 Mini workflow. Click Run, write your prompt, optionally upload reference images, and generate. Floyo handles the ComfyUI environment and API connection. No Dreamina account, no ByteDance API key, no local install.
ByteDance's Seed research team (the same company behind TikTok). Seedance 2.0 launched February 2026 and topped the Artificial Analysis Video Arena. Seedance 2.0 Mini was added as the lightweight tier in June 2026. Seedance 2.5 was announced June 23, 2026 at the Volcano Engine FORCE conference with an early July launch.
Mini is 2x faster and 50% cheaper. Standard outputs up to 2K resolution with tighter lip-sync and maximum visual fidelity. Mini outputs at 480p/720p. Both share the same Dual-Branch DiT architecture, the same input modes, and the same camera/character capabilities. Use Mini for volume work and iteration. Use Standard for final delivery.
Seedance 2.0 Mini is a closed-source API model with native audio, multi-shot direction, and character consistency. Wan 2.7 is open-source with thinking mode, up to 4K resolution, and LoRA training support but no native audio. Mini is faster for iteration. Wan 2.7 is free to run with more customization options. Both are available on Floyo.
Yes. Floyo runs ComfyUI, which lets you chain multiple models. Generate a character with Nano Banana, animate with Seedance 2.0 Mini, add custom narration with ElevenLabs or Fish Audio S2, upscale with Topaz Video AI. Or prototype with Mini and promote winners to standard Seedance 2.0 for final delivery.
Yes. Generate with Seedance 2.0 Mini at 720p, then run the output through Topaz Video AI on Floyo to upscale to 1080p or higher. This two-step workflow gives you Mini's speed and cost during iteration, then production resolution for the final delivery. Both models chain in the same ComfyUI pipeline.
16:9 (widescreen), 9:16 (vertical/TikTok/Reels), 1:1 (square/feed), 4:3 (classic), and 3:4 (portrait). Output exports as ready-to-post MP4. Generate in the format you are publishing in.
Try Seedance 2.0 Mini on Floyo
ByteDance's high-volume video tier. Same Seedance 2.0 architecture at half the cost. Cinematic multi-shot, native audio, camera control, and character consistency. Run it in your browser.
Try Seedance Mini Now → Browse All ModelsRelated Reading
Film and Animation Workflows on Floyo
AI Ad Creatives for Social and Web
Last updated: July 2026. Specs from ByteDance Seed official materials, Volcano Engine FORCE conference (June 23, 2026), Artificial Analysis Video Arena, WaveSpeedAI API documentation, Atlas Cloud performance review, Pexo explainer, SeedanceTips pricing guide, and ZenCreator Seedance guide.
Run Seedance 2.0 Mini through ComfyUI on Floyo. ByteDance's lightweight video tier: same Seedance 2.0 Dual-Branch DiT at 2x the speed and 50% the cost. Multi-shot sequencing, native audio, camera control, and character consistency at 480p/720p. Free to try.