Pika Labs API

The Pika Labs API lets developers generate short video clips from text prompts or still images programmatically. It supports text-to-video and image-to-video workflows at 720p or 1080p resolution.

Pika API pricing runs approximately $0.20 per clip at 720p and $0.45 per clip at 1080p for 5-second generations. Pricing may vary by inference provider and resolution settings.

Yes. Pika 2.2 models are available through fal.ai endpoints. Pika is not available on Wireflow, so there you would call a different video model, such as Gemini Omni Flash 1.1, through the same REST interface.

The Pika API outputs MP4 video files. Clips range from 5 to 10 seconds depending on the generation mode, with support for seven different aspect ratios including 16:9, 9:16, and 1:1.

Pika focuses on fast, affordable short clips. Kling 3 Pro offers longer duration and higher motion fidelity. Runway Gen-3 targets cinematic quality. Pika and Runway are not available on Wireflow; there you can test Kling and other catalog video models through one endpoint.

Pika's inference provider enforces a limit of approximately 20 generations per minute. Wireflow does not call Pika; for the video models it does run, its async job queue handles rate limiting automatically, queuing excess requests and retrying as capacity opens.

Yes. The image-to-video endpoint accepts a source image URL and a motion prompt. The API animates the image with camera movement and subject motion based on your prompt description.

Pikaffects (style transfer effects), lip sync, sound effects, expand canvas, and modify region are currently only available in the Pika web interface and cannot be accessed through the API.