Pony V7

Beyond imagination. Beyond V6. The new standard in AI character art.

Pony the hero

We are getting ready for V7 release - interested in early access? Please reach out at [email protected]

Pony V7

Pony V7 is a powerful, unconstrained foundational model built for character-centric image generation. It excels in creating high-quality visuals across diverse artistic styles—from stylized illustrations and fantasy creatures to anthropomorphic beings and cartoon characters—all through intuitive natural language prompts.

Its versatility makes it an excellent all-in-one model for creators seeking detailed, expressive, and visually coherent character outputs without switching between models.

Core Capabilities

At the heart of V7 is the AuraFlow architecture, which brings notable improvements in coherence, visual fidelity, and stylistic control. The model is fine-tuned with a strong emphasis on anatomy, facial expressions, and dynamic posing, resulting in more natural and accurate character renderings.

Prompt comprehension has been substantially improved—V7 better understands spatial relationships, object descriptions, and compositional cues. It produces more coherent scenes with stronger handling of backgrounds, props, and fine anatomical details such as hands and facial features, all while maintaining close alignment with the original prompt.

What Pony V7 Does Well

Pony generates high-quality, character-focused images in a wide range of visual styles. It supports native resolutions up to 1536×1536 pixels and is designed to follow detailed natural language prompts for lighting, composition, and visual style.

It handles anatomy with improved accuracy, including traditionally difficult areas like hands, feet, and facial expressions. Backgrounds, props, and secondary elements are rendered with better spatial consistency, resulting in more coherent scenes.

Training Data

V7 was trained on approximately 8.5 million aesthetically curated images, balanced across a wide variety of domains. Each image was paired with high-quality, descriptive natural language captions covering both content and style, ensuring strong alignment between prompts and outputs. The dataset was further enriched with metadata, including aesthetic scores, style classifications, and a Content Rating to support content filtering and generation control.

Licensing

Pony V7 is released under the Pony License, permitting free use for individuals, hobbyists, and small businesses, with no restrictions on content usage. Commercial entities with annual revenues exceeding $1 million USD, institutional funding exceeding $1 million USD, or those providing paid inference services should contact [email protected] for commercial licensing terms.

Access the complete Pony License terms here.

Subscribe to our Newsletter

Stay in the loop with exclusive PonyV7 and Fictional.ai updates, event invitations, and special offers delivered straight to your inbox.