As of May 5, 2026, both Seedance 2.0 and Seedance 2.0 Fast are live on Teno. Both tiers support native audio-video generation, start+end-frame image-to-video, and the same prompt grammar, with Fast available for quick drafts and standard reserved for higher-confidence clips. Existing accounts get access today; no plan upgrade is required.
Key Takeaways
- Both tiers ship today. Seedance 2.0 Fast is available for quick drafts, while standard Seedance 2.0 is ready for higher-confidence clips.
- Native audio-video is the headline. ByteDance says Seedance 2.0 uses a unified audio-video architecture for 4-15 second output (ByteDance Seed, 2026).
- The launch page is the quick-start. Use this post for availability and pricing; use the Seedance 2.0 ad-creative guide for prompt examples.
This post is the launch summary: what shipped on Teno, what it costs, who should try Fast first, and when to keep a Veo or Sora workflow in place. For worked transition examples and the fast-vs-standard quality breakdown, use our Seedance 2.0 ad-creative guide.
What's new in Seedance 2.0?
Seedance 2.0 is ByteDance's next-generation video model, announced on February 12, 2026 with four input modalities: text, image, audio, and video (ByteDance Seed, 2026). On Teno today, the ad-team surface is simpler: prompt, start image, optional end image, duration from 4-15 seconds, aspect ratio, audio toggle, and seed.
The four launch-day capabilities that actually change ad-creative pipelines:
- Native audio-video joint generation. Sound and motion come out of one prompt instead of a TTS-then-align pipeline. That's the default for UGC-style or talking-head clips.
- Start+end-frame image-to-video. Anchor both the first and last frames; the model interpolates the motion between them. This is the shortest path to before/after, transformation, and product-reveal beats.
- Multimodal references under the hood. ByteDance documents up to 9 images, 3 video clips, and 3 audio clips as model references; Teno exposes the smaller set ad teams use most often (arXiv:2604.14148, 2026).
- Multilingual audio out of the box. German, French, Japanese, and Spanish UGC variants render audio natively, which removes two hand-offs in cross-border creative pipelines.
- Two tiers, one model family. Standard and Fast share weights and controls; the difference is latency-and-price calibration, not a separately distilled smaller model.
How much does Seedance 2.0 cost on Teno?
At 720p with audio on, standard Seedance 2.0 costs 20 credits per 5-second clip on Teno, while Seedance 2.0 Fast costs 16 credits for the same output. That makes Fast 20% cheaper than standard, and standard about 23% cheaper than Veo 3.1 at 26 credits per 5-second 720p clip with audio on, based on Teno's May 2026 model catalog.
The math at 720p, by clip length:
| Tier | 5s | 8s | 12s |
|---|---|---|---|
| Seedance 2.0 | 20 credits | 32 credits | 48 credits |
| Seedance 2.0 Fast | 16 credits | 26 credits | 39 credits |
| Veo 3.1 | 26 credits | 41 credits | 62 credits |
For paid users, longer durations and 1080p output follow Teno's normal model credit schedule. Live credit-to-USD conversion depends on your active plan; see pricing for the exact rate.
When should you reach for Seedance 2.0 over Veo 3.1 or Sora 2 Pro?
For weekly ad testing, Seedance 2.0 should be the default when you need native audio, start+end-frame control, or lower credit cost in the same batch. In our early May 2026 tests across 14 product categories, about 38% of manually storyboarded ad concepts could start as a Seedance 2.0 start+end-frame generation in under five minutes per variant.
So when should you keep the older workflow? Use this launch-day map:
- UGC and founder-cam. Default to Seedance 2.0 with audio on. Fewer pipeline stitches, lower cost per variant.
- Product reveals and transformations. Seedance 2.0 with start+end frames. Veo 3.1's I2V doesn't expose end-frame anchoring; Sora 2 Pro's doesn't either.
- Photoreal stills-into-motion hero shot. Sora 2 Pro still wins when the seed is a polished still and the prompt is camera moves only.
- Long establishing shots (10-15s slow pulls). Kling O3 Pro is still the right pick on cinematic pull-backs.
- Talking-head ads where lip precision is the headline metric. Veo 3.1 keeps a narrow edge on tight single-shot sync.
For the full comparison matrix, the in-depth fast-vs-standard breakdown, and the worked transition examples (cleaning, skincare, apparel), keep going in the Seedance 2.0 ad-creative guide.
How to use Seedance 2.0 on Teno today
As of the May 5 launch, Seedance 2.0 is in Product Video and Asset Generator with duration from 4-15 seconds, aspect-ratio controls, optional audio, and optional end-frame input. Open either flow, pick Seedance 2.0 or Seedance 2.0 Fast from the model dropdown, then write a prompt or attach a start image for image-to-video.
If your pipeline starts from still frames, create the start and end images in your image flow of choice (for example Product Photo or Asset Generator if you need GPT Image 2), then run image-to-video with both frames in Product Video or Asset Generator — both expose the same Seedance tiers. Want the fastest test? Start a Teno account; the Fast tier is available without a credit card. For where this fits in the broader paid-creative stack, see our AI ad creative platform guide. For the multi-model studio that now hosts Seedance 2.0 alongside Veo, Kling, Sora, and the 7-image stack, see the Asset Generator launch post.
Sources
- ByteDance Seed, Official launch of Seedance 2.0, retrieved 2026-05-05.
- Team Seedance+, Seedance 2.0: Advancing Video Generation for World Complexity, arXiv:2604.14148, retrieved 2026-05-05.
- Teno model catalog, May 2026 (credit calibration and 1080p multiplier). Internal documentation; live credit-to-USD conversion reflected on /pricing.
