Kling 3 Pro - Kyma API

Overview

This model generates video from text prompts or input images via an image_url parameter. It delivers sharper detail and smoother motion than its predecessor, with a focus on photorealistic human subjects and cinematic framing. On Kyma, requests route through an OpenAI-compatible endpoint with automatic failover to maintain reliability. Prompt caching is supported, billing repeated prompt prefixes at this model’s cached input rate. Every response returns the exact compute cost in usage.cost and identifies the active model in the X-Kyma-Model header. The model operates on a premium cost tier and runs at a slower inference speed compared to fast-tier alternatives. It does not natively generate audio; for synchronized soundtracks, you must route to the dedicated audio variant.

Specs

Field	Value
Model ID	`kling-3-pro`
Best for	Premium cinematic clips, character/face shots, hero brand video
Context window	0
Input modalities	Text, Image
Output modalities	Video
Tool calling	Yes
Structured outputs	Yes
Prompt caching	Yes
Speed	Slow
Cost band	Premium
Release stage	Stable

Pricing

	Per 1M tokens
Input	$0.20
Output	$0.60

Use this when

Cinematic Character Shots — Generate high-fidelity video clips focused on realistic human faces and smooth motion.
Brand Hero Videos — Produce premium marketing assets from text prompts or reference images.
Image-to-Video Conversion — Animate static reference frames into continuous, photorealistic video sequences.
Premium Visual Prototyping — Test high-quality video concepts before committing to full production pipelines.

Not ideal for

Do not use this model for real-time applications, low-budget batch generation, or workflows that require native audio synthesis.

Pick something else when

You need faster inference speeds: use seedance-2-fast or veo-3-fast.
You require native audio generation: use kling-3-pro-audio.
You are generating lower-resolution drafts: use hailuo-02-512p or hailuo-02-768p.
You need a cheaper alternative for bulk testing: use kling-2.5-pro.

Example

from openai import OpenAI

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

response = client.chat.completions.create(
    model="kling-3-pro",
    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=frontier-open"

FAQ

How do I pass an image to generate video? Include the image URL in the image_url field of your request payload alongside your text prompt. Does this model generate audio with the video? No, it outputs silent video only. Route to kling-3-pro-audio if you need synchronized soundtracks. How does prompt caching affect my costs? Repeated prompt prefixes are billed at this model’s cached input rate, which reduces expenses for iterative generation workflows.

​Overview

​Specs

​Pricing

​Use this when

​Not ideal for

​Pick something else when

​Example

​Agent query example

​FAQ