Veo 3 Fast - Kyma API

Overview

Veo 3 Fast generates 720p video from text or image prompts. It operates without audio output and sits at the balanced cost and speed tier within the Veo family. The endpoint accepts text and image inputs and returns video files, making it suitable for iterative visual development. On Kyma, requests route through an OpenAI-compatible base URL using a single API key. The platform applies automatic failover if a serving path degrades, and responses include the exact generation cost in the usage.cost field. Prompt prefixes are cached automatically, billing repeated inputs at this model’s cached input rate. The model does not support reasoning, vision understanding, or structured outputs. It is strictly a generation endpoint. Output length is billed per second, and the model is intended for draft-quality or standard-resolution needs rather than final broadcast assets.

Specs

Field	Value
Model ID	`veo-3-fast`
Best for	Default Veo tier — balanced quality and speed
Context window	0
Input modalities	Text, Image
Output modalities	Video
Tool calling	Yes
Structured outputs	Yes
Prompt caching	Yes
Speed	Medium
Cost band	Balanced
Release stage	Stable

Pricing

	Per 1M tokens
Input	$0.20
Output	$0.60

Use this when

Social Media Drafts — Generate quick 720p clips for testing visual concepts before final production.
High-Volume Iteration — Run rapid prompt variations to explore visual directions without high latency.
Storyboard Visualization — Convert image references or text prompts into short video sequences for planning.
Automated Content Pipelines — Integrate into batch workflows where cost efficiency and medium speed are prioritized.

Not ideal for

Do not use this model for final broadcast assets, projects requiring synchronized audio, or workflows demanding 1080p resolution.

Pick something else when

You need synchronized audio in the output: use veo-3.
You require 1080p resolution: use hailuo-02-1080p.
You need faster generation for real-time previews: use seedance-2-fast.
You need higher fidelity for final production: use kling-3-pro.

Example

from openai import OpenAI

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

response = client.chat.completions.create(
    model="veo-3-fast",
    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

Does this model generate audio alongside the video? No, Veo 3 Fast outputs video only. Use the standard Veo 3 tier if audio is required. How does Kyma handle request routing and billing? Every request uses automatic failover to maintain uptime, and the exact generation cost is returned in the usage.cost field. Cached prompt prefixes are billed at this model’s cached input rate. Can I use this with my existing OpenAI client libraries? Yes, Kyma provides an OpenAI-compatible base URL and accepts a single API key across all models.

​Overview

​Specs

​Pricing

​Use this when

​Not ideal for

​Pick something else when

​Example

​Agent query example

​FAQ