Sub-second speed and ultra-low cost without sacrificing quality.
Cost, Speed and Quality
Just $0.005 and 1 second per image.
Prompt Adherence
Exact prompt adherence for precise results.
Text Rendering
Highly controlled text generation.
Prompting Guidelines
Start with the Core: Combine the three basic elements for an instant result.
Subject: The main focus (e.g.,
a robotic fox)Scene: The environment (e.g.,
in a neon-lit forest)Style: The overall aesthetic (e.g.,
cyberpunk, digital art)
Iterate in Real-Time: Don't waste time on a "perfect" prompt. Get your first idea out instantly. See the result, change the prompt, and generate again. Your workflow is now a live conversation, not a waiting game.
Explore Freely: For $0.005 and 1 second per image, you're encouraged to experiment and explore. Try wild ideas. See your result in less than a second. Don't like it? Instantly swap one part.
Build on Your Core: Start with your
Subject,Scene, andStyle, but add specific layers of detail.Direct the Camera: Add "lens usage" to guide the shot.
Control the Mood: Use "atmosphere words" and lighting commands.
Add Detail Modifiers: Pinpoint the small things that matter.
Be Direct to Be Efficient: A clear prompt is a more efficient one. The more directly you can describe your vision, the faster you'll get your final, production-ready image.
P-Image 1.0 delivers quality on par with SoTA models like Flux 1.1 Pro but is 8x cheaper and 4.8x faster. P-Image-Edit 1.0 introduces powerful multi-image editing, outperforming all competitors in speed, cost, and capability.
Acknowledgments
The launch of P-Image 1.0 is a milestone for real-time, high-quality image generation. But it is not a journey we took alone.
We want to acknowledge the teams whose work defined the state-of-the-art, introduced the proper techniques, and set the high bar for quality that we aimed to match.
Black Forest Labs (BFL): Their work on the Flux series
Alibaba: Their work on Qwen Image, Qwen Image Edit and Wan Video
The Research Community: For the fantastic work on countless open-source initiatives and optimisation algorithms.
Pruna is a member of the Coalition for Sustainable AI.
Pruna co-authored "Key challenges for the environmental performance of AI".
Code Carbon is integrated into Pruna.