Dreamina API v1 now supports the CA (Canada) region with access to Seedance 2.0 — ByteDance’s latest and most capable video generation model. Also new: 1080p resolution, video upscale (2x super resolution), and Omni Reference mode for interleaved text and image references.
All videos below were upscaled to 2x resolution using POST /videos/upscale.
Omni Reference — interleaved text and image references
Upload reference images via POST /assets, then use @imageN placeholders in your prompt to reference them.
@image1
@image2
@image3
curl --location 'https://api.useapi.net/v1/dreamina/videos' \
--header 'Authorization: Bearer user:12345-…' \
--form 'prompt="Cinematic, photorealistic tracking shot. A @image1 rides the @image2 along a sandy, rocky beach, cutting thru the waves. @image3 looms in the foggy background with crashing ocean waves. @image1 lets go of the handlebars, spreads her arms wide into the wind, and shouts joyfully with perfect lip-sync: '\''Whoo hoo! I'\''m flying!'\'' Dynamic motion, wind effects, highly detailed."' \
--form 'ratio="16:9"' \
--form 'duration="15"' \
--form 'omni_1_imageRef="CA:[email protected]:w1088:h1920:s1800540-uri:…"' \
--form 'omni_2_imageRef="CA:[email protected]:w2560:h1440:s181878-uri:…"' \
--form 'omni_3_imageRef="CA:[email protected]:w2848:h1600:s442898-uri:…"'
First Frame — image-to-video with lip-sync
curl --location 'https://api.useapi.net/v1/dreamina/videos' \
--header 'Authorization: Bearer user:12345-…' \
--form 'prompt="The giant red, winged demonic monster looming over the warrior suddenly stops acting menacing. It drops its clawed hands, sighs, and speaks in a deep, terrifying, roaring monster voice with perfect lip-sync: '\''Hold up, dude! I'\''m literally just walking to the market to get some wine!'\'' The massive demon then comically slumps its shoulders and sits down in the dusty ruins. Ignoring the demon, the warrior in the red cape suddenly turns his head to face the camera directly. Looking straight at the viewer, he proudly raises his fist and shouts in a booming voice with perfect lip-sync: '\''Glory to Rome!'\'' High fantasy meets mundane comedy, cinematic motion, realistic dust particles."' \
--form 'model="seedance-2.0"' \
--form 'duration="15"' \
--form 'firstFrameRef="CA:[email protected]:w1376:h768:s695678-uri:…"'
Text-to-video — prompt only with lip-sync
curl --location 'https://api.useapi.net/v1/dreamina/videos' \
--header 'Authorization: Bearer user:12345-…' \
--form 'prompt="TikTok POV selfie-camera angle. A stylish, trendy Gen-Z girl stands outside an aesthetic pastel coffee shop on a sunny street, holding up a plastic iced coffee cup. She looks directly into the camera lens with a deadpan, slightly traumatized expression and speaks in a fast, sarcastic, comedic voice with perfect lip-sync: '\''I just paid twelve dollars for an oat milk latte... and the barista looked me dead in the eyes and wrote Gargoyle on my cup. My name is Chloe.'\'' She sighs, breaks eye contact, and takes a slow, aggressive sip of her iced coffee. Modern influencer aesthetic, bright natural lighting, highly detailed, photorealistic."' \
--form 'model="seedance-2.0"' \
--form 'ratio="9:16"' \
--form 'duration="15"'