UlazAI - AI Image & Video Tools
Seedream AI Image-to-Video Generator
Animate Seedream images into short videos with V1 Pro or V1 Lite. Compare duration, resolution and camera controls before generation.
Seedream 3.0 (Images)
Ultra‑fast image generation
Generate images with multiple aspect ratios and review text rendering before use.
Clear prompt controls
Guidance scale, size presets, safety checker, and optional seed for reproducibility.
Transparent pricing
8 credits per image. No hidden fees. Results render inline with quick previews.
Need 4K still images first?
Use the dedicated Seedream 4.5 image generator page for 4K text-to-image, image editing, aspect ratios, and quality modes. Then return here when you want to animate the output into video.
4K Quality Images
Generate ultra-high-resolution images with incredible detail. Choose from Standard, HD, or 4K quality modes.
AI Image Editing
Edit existing images with text prompts. Change backgrounds, add elements, or modify styles - all with natural language.
Fast Processing
Uses asynchronous processing and status polling for supported 4K image jobs.
V1 Pro vs V1 Lite (Image‑to‑Video)
V1 Pro — Higher control
Pick a resolution (480p/720p/1080p), set 3–12s duration, and optionally fix the camera. Designed for quality marketing shots.
- • Inputs: prompt, image_url, resolution, duration, camera_fixed
- • Credits per second by resolution (6/10/16)
V1 Lite — Fast iteration
A streamlined flow for quick concepts. Set 3–12s duration; pricing starts at 20 credits (3s) and increases by 5 per extra second.
- • Inputs: prompt, image_url, duration, resolution
- • Credits: 3s=20, then +5 per extra second
How it works
- Write a clear prompt (use the Enhance button for better motion and detail).
- For video, paste an image URL and choose V1 Pro or V1 Lite with duration and resolution.
- Submit. We process async via webhook + polling; results render inline when ready.
FAQ
Do I need to host the input image?
Yes, provide a public https URL (png/jpg/webp ≤10 MB). For privacy‑sensitive assets, use temporary links.
Can I disable the safety checker?
Image Studio enables it by default; via API you can toggle with enable_safety_checker.
What happens on failures?
Failures are marked in the app and credits are automatically refunded to your balance.
API Endpoints
Submit Seedream image or video tasks via REST. Authenticate with session or API token.
/api/v1/seedream/generate/Image (Seedream 3.0)
curl -X POST https://ulazai.com/api/v1/seedream/generate/ \
-H 'Content-Type: application/json' \
-b 'sessionid=YOUR_SESSION' \
-d '{
"generation_type": "image",
"prompt": "Modern product photo, studio, soft edge light",
"image_size": "square_hd",
"guidance_scale": 2.5,
"enable_safety_checker": true
}'
Video — V1 Pro (Image‑to‑Video)
curl -X POST https://ulazai.com/api/v1/seedream/generate/ \
-H 'Content-Type: application/json' \
-b 'sessionid=YOUR_SESSION' \
-d '{
"generation_type": "video",
"prompt": "Parallax push-in, subtle particles over logo plate",
"video_model": "bytedance/v1-pro-image-to-video",
"input_image_url": "https://example.com/still.jpg",
"resolution": "720p",
"duration": 6,
"camera_fixed": false
}'
Video — V1 Lite (Image‑to‑Video)
curl -X POST https://ulazai.com/api/v1/seedream/generate/ \
-H 'Content-Type: application/json' \
-b 'sessionid=YOUR_SESSION' \
-d '{
"generation_type": "video",
"prompt": "Slow reveal, ambient dust, gentle camera push",
"video_model": "bytedance/v1-lite-image-to-video",
"input_image_url": "https://example.com/still.png",
"resolution": "720p",
"duration": 5,
"camera_fixed": false
}'
/seedream/status/<generation_id>/curl https://ulazai.com/seedream/status/UUID/ -b 'sessionid=YOUR_SESSION'
/api/v1/seedream/enhance/curl -X POST https://ulazai.com/api/v1/seedream/enhance/ \
-H 'Content-Type: application/json' \
-b 'sessionid=YOUR_SESSION' \
-d '{"prompt":"Hero product on matte plate, cinematic lighting","generation_type":"video"}'
Webhook callbacks finalize tasks; backend polling acts as fallback. Credits are auto‑refunded on failure. See the Seedream API docs or the full API docs for authentication options.
Use the Seedream 4.5 page for image generation
Generate or edit a Seedream 4.5 image, then return here when you want to animate that image.
Use Seedream to create the still, then choose a video model for motion
Seedream generates and edits images in UlazAI. Build a clean source image first; if you need image-to-video, carry that still into a listed video model instead of assuming Seedream itself supplies motion.
Check composition, text and product accuracy before animation. Small image errors become more obvious once frames move.