This model is deprecated. No direct replacement is recommended yet.
Video
Pika 2.2 T2V
Use it ↗Pika 2.2 T2V by Pika Labs generates cinematic video from text prompts at 720p or 1080p, in 5 or 10 second clips, across landscape, square, or portrait formats.
Pika 2.2 T2V is a closed-source model (Feb 2025) that generates 1080p video from text. It prioritizes cinematic quality and physics over speed, making it a solid choice for high-fidelity concept shots. While not positioned against SOTA models like Kling 2.6, it is distinguished by Pika’s unique “Pikaffects” system for physical interactions. It accepts text-only input to produce clips up to 10 seconds long, offering a reliable option for foundational video creation when no source image is needed.