Turnaround
Faster returns
Positioned by ByteDance for low-latency use cases
Generate videos with reference materials using the Speed Edition of Seedance 2.0.
Seedance 2.0 image-to-video fast version model.
Seedance 2.0 fast is a new-generation multimodal video creation model launched by the Doubao Large Model Team. It inherits the core functions and advantages of the Seedance 2.0 model and features faster generation speed. This is a version of the text-to-video model.
Official lineup · Seedance 2.0 Fast
Faster turnaround, ~20% cheaper per token — the efficiency pick for drafts and previews.
Fast shares the Standard version's full task set — text-to-video, image-to-video, all-modal references, editing, and extension — at 480P / 720P output with faster returns and lower unit pricing. Draft fast with Fast, then finish at 1080P / 4K on Standard.
Turnaround
Faster returns
Positioned by ByteDance for low-latency use cases
Unit price
~20% off
Token price at roughly 80% of Standard
Output
480P / 720P
Covers short-form and preview scenarios
Task types
All 6 covered
T2V / I2V / reference / edit / extend
One model family: Fast owns "fast and many", Standard owns "refined and final".
Reference prices for a 5s 16:9 clip; token-based billing, successful tasks only. Always refer to live rates on Volcano Ark and the APIPod pricing page.
| Unit: USD per 1M tokens (no video reference input) | 5s 16:9 per-clip reference price |
|---|---|
| Fast · 480P | $5.1 / 5s ≈ $0.26 |
| Fast · 720P | $5.1 / 5s ≈ $0.55 |
| Standard · 480P | $6.4 / 5s ≈ $0.32 |
| Standard · 720P | $6.4 / 5s ≈ $0.69 |
Fast matches Standard on task capability per resolution tier; 1080P / 4K remain Standard-only.
Fast isn't a stripped-down build — it's a deliberate speed/cost trade-off.
T2V, I2V (first/first-last frame), all-modal reference generation, editing, and extension — same as Standard.
Audio-video joint generation with stereo output — voice, SFX, and music tracks.
Inherits the Seedance 2.0 multimodal reference mechanism; exact input limits follow your live channel.
Output tops out at 480P / 720P; switch model to seedance-2.0 whenever you need 1080P / 4K.
Demos from the official Seedance 2.0 launch blog — Fast shares the same task system as Standard and produces this class of content with faster returns.
R2V · Storyboard to film
Upload a storyboard plus character, scene, and prop references to get a finished 15s short in one shot — with Fast, this kind of rapid validation costs almost nothing.
Official prompt
Follow the storyboard in @image1 for shot division, framing, camera movement, visuals, and copy; the character is @image2, the scene is @image3, the prop is @image4. Create a 15-second healing short.
T2V · Multi-shot story
The model plans its own shot language, turning one prompt into a complete narrative arc — perfect for validating creative directions in batches.
Official prompt
A Lunar New Year family album comes alive: portraits animate as the camera pans past — grandparents, parents, and kids each perform a signature gesture. Lanterns and couplets light up as the frame converges into a lively family photo, laughter in sync.
T2V · Native audio
Rain, thunder, and the shockwave of clashing blades land exactly on the frame, in native stereo.
Official prompt
A wuxia duel in a bamboo forest: a white-robed swordsman faces a coir-clad blade fighter. Slow push-in, focus shifting between raindrops and sword hilts — only rain is audible. Thunder cracks, both charge; the clash drops into extreme slow motion, rain rings blasted off the blades, bamboo leaves sliced mid-air.
Generate multiple drafts cheaply to lock shot language and pacing.
Keep the strong directions; fine-tune prompts and reference packs.
Point the winning prompt at seedance-2.0 for the 1080P / 4K master.
Just change the model field — auth, billing, and observability carry over.
Example request
Same API as Standard — set model to seedance-2.0-fast.
Parameter names may vary by provider channel and model revision. Use your live APIPod model docs as the source of truth.
curl -X POST https://api.apipod.ai/v1/videos/generations \
-H "Authorization: Bearer $APIPOD_API_KEY" \
-H "Content-Type: application/json" \
-d {
"model": "seedance-2.0-fast",
"prompt": "City nightscape timelapse with flowing neon lights and upbeat electronic music.",
"duration": 5,
"resolution": "720p",
"aspect_ratio": "16:9",
"generate_audio": true
}Fast is cut from the same cloth as Standard; the real differences are the 480P/720P resolution cap and the speed/cost trade-off. Within those tiers quality is the same family tier — ByteDance positions it for low-latency, cost-sensitive workloads.
No. 1080P and 4K (10-bit) are Standard-only. For high-resolution masters, switch model to seedance-2.0.
Yes. Fast supports audio-video joint generation with native stereo output synced to the visuals.
Creative drafts, high-concurrency previews, prompt A/B tests, and b-roll/backdrop assets — anything that needs to be fast and plentiful. Hand final delivery to Standard.
One integration on APIPod — switching between Fast and Standard is just a model field away.