Cinematic Adventure
16:9Man walked into winter cave with polar bear
Hailuo Pro
ID:
hailuo/proHailuo Pro video generation API by MiniMax — fast text-to-video and image-to-video with strong motion quality and character consistency.
Text to videoImage to video
Input
OutputView all
Output will appear here
Fixed
$0.31
cURL
curl https://api.runbase.net/v1/runs \
-H "Authorization: Bearer $RUNBASE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "hailuo/pro",
"input": {
"prompt": "A cinematic product photo of a ceramic lamp",
"aspect_ratio": "1:1",
"resolution": "1K"
}
}'Examples
Trail Action
1:1A mountain biker navigating a forest trail at speed, launching over a jump, dust and dirt flying, golden backlight through the trees
Sports Cinematography
16:9Athlete running powerfully on beach, dynamic camera movement tracking the runner, waves and sunset in motion, Hollywood movie cinematography, professional sports filming, inspiring atmosphere
Overview
Hailuo Pro is built on MiniMax's Hailuo 02 (Pro tier), released in 2025. It focuses on two things: motion quality and character consistency. The model produces natural, physically plausible movement — fluid body mechanics, realistic weight, and coherent object trajectories. On Runbase, this model exposes a deliberately minimal interface: just a text prompt and an optional source image. Resolution, duration, and aspect ratio aren't exposed as parameters here — the integration uses the provider's defaults.
Use cases
Social media content where natural motion and character consistency matter. Quick text-to-video generation with minimal configuration. Image-to-video for animating portraits, product shots, or illustrations. Action and sports content where fluid body mechanics are critical.
Inputs
All parameters are passed in the input object of the run request.
| Parameter | Required | Description |
|---|---|---|
| prompt | Yes | Text description (1–1500 chars) |
| image_urls | No | Source image (max 1, max 10 MB) for image-to-video |
Prompt tips
Keep prompts concise and action-focused
With a 1500-character limit, every word counts. Lead with the subject and action: "A dancer spinning on a rooftop at sunset" works better than elaborate scene-setting. The model excels at interpreting brief, vivid descriptions.
Leverage the model's motion strengths
Hailuo Pro is built for natural movement. Prompts involving running, dancing, sports, flowing fabric, or animal locomotion play to its strengths. Describe the type of motion you want — "slow graceful turn" vs. "explosive sprint" — and the model handles the physics.
Limitations
- Short prompt limit (1500 characters) — significantly less than most competitors
- Resolution is not exposed as a parameter on Runbase — the provider default is used
- Duration is not exposed as a parameter on Runbase — the provider default is used
- No aspect ratio parameter — framing follows the source image or provider default
- Only one source image supported for image-to-video
FAQ
Why are there so few parameters?
This Runbase integration keeps the interface minimal — you send a prompt and an optional source image, and resolution and duration fall back to the provider's defaults. This simplifies the workflow at the cost of fine-grained control.
Can I control the output resolution or aspect ratio?
Not on Runbase — this integration doesn't expose resolution, aspect ratio, or duration parameters, so the provider's defaults apply. (The underlying Hailuo 02 model does support duration and resolution settings; they're just not surfaced here.)
How does image-to-video work?
Provide a single image via image_urls and the model uses it as the starting frame, generating motion from that visual. The image must be under 10 MB.