Pruna AI provides a powerful collection of image and video generation models designed for high-performance creative AI workflows. The Pruna model suite covers text-to-image generation, image editing, image upscaling, LoRA-based customization, image-to-video generation, text-to-video generation, and avatar video creation.
These models are built for developers, creators, and AI applications that need fast, flexible, and scalable visual generation capabilities.
P-Image Series - Image Generation and Editing
pruna-ai/p-image/text-to-image
Pruna's core text-to-image generation model, designed to create high-quality images from natural language prompts.
Key features include:
Text-to-Image Generation: Creates images directly from detailed text prompts
Creative Control: Supports diverse styles, subjects, compositions, and visual concepts
Fast Visual Creation: Suitable for product images, concept art, social media assets, and creative design workflows
Application Ready: Easy to integrate into AI image generation platforms and creator tools
pruna-ai/p-image/edit
An image editing model that transforms existing images based on user instructions.
Key features include:
Image-to-Image Editing: Modifies an input image according to a text prompt
Object and Style Changes: Supports visual adjustments, subject transformation, and scene modification
Creative Retouching: Useful for design refinement, character editing, and content enhancement
Prompt-Guided Control: Allows users to describe the desired edit in natural language
pruna-ai/p-image/upscale
An image upscaling model designed to improve image resolution and visual clarity.
Key features include:
Image Upscaling: Enhances low-resolution images into higher-resolution outputs
Detail Enhancement: Improves sharpness, texture, and visual quality
Restoration Support: Useful for old photos, compressed images, and generated image refinement
Production Utility: Suitable for thumbnails, marketing assets, and high-quality visual exports
LoRA and Training Models
pruna-ai/p-image/text-to-image-lora
A text-to-image generation model with LoRA input support, enabling customized visual generation using user-provided LoRA styles or subjects.
Key features include:
LoRA Input Support: Generates images using custom LoRA models
Personalized Output: Supports specific styles, characters, products, or visual identities
Flexible Generation: Combines prompt-based generation with LoRA customization
Creator Friendly: Ideal for consistent character generation, brand assets, and stylized content
pruna-ai/p-image/text-to-image-trainer
A LoRA training model for creating customized text-to-image LoRA models.
Key features include:
LoRA Training: Trains custom LoRA models from user-provided image datasets
Custom Style Learning: Learns specific characters, products, styles, or visual patterns
Personalized Image Generation: Enables downstream text-to-image generation with trained LoRA models
Workflow Integration: Designed for users who need repeatable and consistent custom image outputs
pruna-ai/p-image/edit-lora
An image editing model with LoRA input support, enabling customized image edits using LoRA-based personalization.
Key features include:
LoRA-Based Image Editing: Applies custom LoRA styles or identities during image editing
Consistent Subject Control: Helps preserve specific characters, products, or visual styles
Prompt-Guided Modification: Edits images according to natural language instructions
Advanced Creative Workflow: Useful for personalized image retouching and branded visual generation
pruna-ai/p-image/edit-trainer
A training model for building LoRA models used in customized image editing workflows.
Key features include:
Edit LoRA Training: Trains LoRA models for personalized image editing
Custom Identity Learning: Supports consistent characters, objects, products, or visual styles
Enhanced Editing Control: Improves the quality and consistency of prompt-based edits
Professional Customization: Suitable for creators and platforms that require reusable editing styles
P-Video Series - Video Generation
pruna-ai/p-video/text-to-video
Pruna's text-to-video generation model, designed to create videos from natural language prompts.
Key features include:
Text-to-Video Generation: Creates video clips directly from text descriptions
Dynamic Motion: Generates scenes with movement, camera motion, and visual continuity
Creative Storytelling: Suitable for short videos, concept previews, social media clips, and visual narratives
Prompt-Based Control: Users can describe characters, scenes, actions, and cinematic style
pruna-ai/p-video/image-to-video
An image-to-video model that animates a static image into a dynamic video sequence.
Key features include:
Image-to-Video Generation: Converts still images into motion videos
Visual Consistency: Preserves the original image subject and visual style
Motion Expansion: Adds natural movement, camera dynamics, and scene animation
Creative Animation: Useful for bringing portraits, product images, illustrations, and generated images to life
pruna-ai/p-video/avatar
An avatar video generation model for creating presenter-style or character-based videos.
Key features include:
Avatar Video Creation: Generates avatar-based video content
Presenter Workflow: Suitable for virtual presenters, AI hosts, product explainers, and social media videos
Character Consistency: Maintains a stable avatar appearance across generated content
Content Production Ready: Helps automate video creation for marketing, education, and digital communication