Hailuo 02 (512p) - Kyma API

Overview

This model generates 512p video clips from text prompts or an input image via image_url. It supports both 6-second and 10-second durations and operates on a per-clip billing model. It is the cheapest video tier available on the platform. On Kyma, it runs through an OpenAI-compatible endpoint using a single API key. New accounts start with a $0.50 credit, and prompt caching bills repeated prefixes at this model’s cached input rate. Requests automatically fail over if a serving path degrades, and the exact cost is returned in the usage.cost field. The model is optimized for cost over speed, so generation times will be slower than higher-tier alternatives. It does not support structured outputs, reasoning traces, or vision input parsing beyond the initial image reference.

Specs

Field	Value
Model ID	`hailuo-02-512p`
Best for	Social shorts, motion mockups, rapid iteration, budget video workflows
Context window	0
Input modalities	Text, Image
Output modalities	Video
Tool calling	Yes
Structured outputs	Yes
Prompt caching	Yes
Speed	Slow
Cost band	Cheap
Release stage	Stable

Pricing

	Per 1M tokens
Input	$0.20
Output	$0.60

Use this when

Social Media Shorts — Generate quick 6s or 10s clips optimized for platforms like TikTok or Instagram.
Rapid Motion Mockups — Produce low-fidelity video drafts to test pacing and composition before high-res rendering.
Budget Video Workflows — Scale automated video generation pipelines where per-clip cost matters more than latency.

Not ideal for

Do not use this model for high-resolution cinematic output, real-time interactive applications, or workflows requiring structured JSON responses.

Pick something else when

You need higher resolution or faster generation times: use hailuo-02-768p or hailuo-02-1080p.
You require premium visual fidelity and complex scene understanding: use kling-2.5-pro or veo-3.
You need audio generation alongside video: use kling-3-pro-audio.

Example

from openai import OpenAI

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="ky-...")

response = client.chat.completions.create(
    model="hailuo-02-512p",
    messages=[{"role": "user", "content": "..."}],
)

Agent query example

Ask the API which models fit, instead of hardcoding an id:

curl "https://kymaapi.com/v1/models?recommended_for=generate-video&tools=true&quality_tier=strong"

FAQ

Can I pass an image to generate a video? Yes, provide the image URL in the image_url field to trigger image-to-video generation. How does prompt caching work for this model? Repeated prompt prefixes are billed at this model’s cached input rate. What happens if the serving endpoint slows down? Kyma automatically reroutes your request to a healthy path via built-in failover, and the X-Kyma-Model header confirms which model actually processed the request.

​Overview

​Specs

​Pricing

​Use this when

​Not ideal for

​Pick something else when

​Example

​Agent query example

​FAQ