Pika 2.5
AI model · Video generation · Text → Video by Pika
Cost-efficient 720p AI video generation from text, images, and keyframes.
A cost-efficient AI video model for generating 5-second 720p clips from text, images, or keyframes. Ideal for creators seeking a balance between quality and credit economy across web and mobile platforms.
Model specs
The full spec card (context, pricing per token, modalities and more) on AntSpace. Sign up free to see the full spec card.
- Languages
- English
Category: Generative AI
- Modalities
- Text, image, keyframes → video
- Resolutions
- 720p at 5 seconds
- Output formats
- MP4
- Access
- Web platform, iOS app, ChatGPT plugin
- Default Resolution
- 720p
- Default Duration
- 5 seconds
- 720p resolution at 5 seconds duration
- 20 credits per generation cost-efficiency
- Supports text, image, and keyframe inputs
- Available on web, iOS, and ChatGPT via MCP
Overview
Pika 2.5 is a cost-efficient AI video generation model designed for creators who need to produce high-quality content while managing credit usage. It generates 5-second video clips at 720p resolution from text prompts, reference images, or specific keyframes. The model is integrated across the Pika ecosystem, including the web platform, the official iOS application, and the Pika MCP plugin for ChatGPT. Pika 2.5 is accessible to all users, with commercial usage rights available to those on paid subscription plans.
Creation Modes
- Text to video generation: Generates video content from natural language descriptions provided by the user.
- Image to video generation: Uses a reference image to define the visual style or the first frame of the generated video.
- Keyframe-guided animation: Directs the animation by defining specific keyframes to guide the motion and composition.
Output Specifications
- 720p Resolution: Produces video at a standard 720p resolution for a balance of quality and speed.
- 5-second duration: Generates clips with a default length of 5 seconds per pass.
- Cost-efficient processing: Operates at a lower credit cost of 20 credits per generation compared to frontier models.
Use cases
- Creators use Pika 2.5 to generate short 5-second social media clips from text prompts at a lower credit cost.
- Designers animate static images into 720p videos by using them as reference frames for the model.
- Video editors use keyframe-guided generation to maintain specific compositions across AI-generated sequences.
Specifications
- Modalities
- Text, image, keyframes → video
- Resolutions
- 720p at 5 seconds
- Output formats
- MP4
- Access
- Web platform, iOS app, ChatGPT plugin
- Default Resolution
- 720p
- Default Duration
- 5 seconds
- Credit Cost
- 20 credits per generation
- Input Modalities
- Text, Image, Keyframes
- Platform Availability
- Web, iOS, ChatGPT Plugin
How it works
- Select the model: Select Pika 2.5 from the model picker in Video Studio or other compatible Pika apps.
- Provide inputs: Enter a text prompt, upload a reference image, or set keyframes to define the desired video content.
- Generate and review: Click generate to produce a 5-second 720p video clip, which consumes 20 credits from the account balance.
Pricing
- Credits per generation
- 20 credits
FAQ
What is Pika 2.5?
Pika 2.5 is a cost-efficient video generation model that creates 5-second clips at 720p resolution from text, images, or keyframes.
How many credits does Pika 2.5 use?
Each generation with Pika 2.5 costs 20 credits on the Pika platform.
Where can I use Pika 2.5?
Pika 2.5 is available through the Pika web interface, the official iOS app, and the Pika MCP plugin for ChatGPT.
Can I use Pika 2.5 for free?
Yes, Pika 2.5 is included in all Pika plans, including the Free tier, though a commercial license requires a paid subscription.
Added by AntSpace from public sources (pika.art).