So… you want to hear a story, eh? One about creation, ambition, and endless possibilities? Haha, well… have I got a story for you!
Pandora is an all-purpose model built to handle a wide range of creative outputs. It aims to blend flexibility, efficiency, and prompt comprehension into a cohesive model.
The model is composed of:
Photographic Content (75%)
Artistic and Anime (25%)
Key Capabilities
Portraits and Characters
Anime and Digital Art
Macro and Detail Work
Painting and Conceptual Art
Current Model Goals
Detailing and Realism: Enhancing overall detail, including skin textures, reflections, and lighting
Faces and Anatomy: Producing more realistic faces, chins, and accurate proportions
Art Styles: Supporting painting, digital art, and anime styles
Photographic Styles: Refining the ability to create photorealistic images with dynamic lighting and rich textures
Faster Generation: Performance optimizations and reduced generation time
Higher Resolution: Improving the ability to generate images with resolutions above 2MP
Negative Prompting (Pro Version): Enhancing negative prompting capabilities for more precise control (currently in progress and showing promising results)
Known Issues
Anime Style: While improved, anime could still be better, it is recommended to pair the model with my Pandora | Anime and Digital Art LoRa
Painting Styles: Painting styles are coming in a future update
Current Enhancements
Better Details: Enhanced textures and lighting across various styles
Photographic Style: Outstanding results with photorealistic images with better look and composition
Better Realism: Realistic scenes, faces, and settings have become more consistent and detailed
Improved Anime: Anime-style images are more polished compared to Flux-Dev or Schnell
Available Versions
Dev Version: Focused on adaptability and versatility.
Pro Version: A more refined version built from the dedistilled version for enhanced performance, including negative prompting and other advanced features. Mainly the use of CFG or Classifier Free Guidance, giving it an edge over the Dev version and allowing negative prompting, but making it harder to master.
Why Beta Testing is Key
Pandora is now in beta, meaning there is still room for growth and refinement. The goal is for it to become a well-rounded, flexible model that adapts to diverse creative needs, from realism and anime to conceptual storytelling.
The pro version uses a Flux dedistilled core, but there is a distinctive dev and pro branches created for different needs. Feedback is crucial at this stage to shape it into the all-purpose model it’s meant to be.
Description
***Pandora-Pro is based on Flux Dedistilled***
Recommended Parameters
CFG: 3-6 (not distilled guidance)
Steps: 40 to 60
Sampler: DPM++ 2M
Scheduler: Beta
With Pandora-Pro Beta, there is a direct relationship between the number of steps and the CFG value used during generation:
CFG × 10 = Total Number of Steps
Beware that higher CFG values will naturally result in longer generation times. However, tthe higher the CFG, the smarter and more context-aware it becomes.
For best results:
• CFG 3 to 4: Balanced detail and precision.
• High CFG (5+): Maximum comprehension, but with longer render times.
Note: This release does not include the t5_xxl, CLIP-L, or VAE files. You will need to download them separately for proper functionality:
Text Encoders : t5_xxl and clip-l (FP8 version recommended).
VAE: Black Forest Labs VAE (ae.safetensors).
FAQ
Comments (10)
First of all, I really love your model – it’s by far the best overall one I’ve tried, so thank you for sharing it! 🙏
I’ve been experimenting with the Pro version and had a question about the text encoders: do you recommend using the CLIP + T5 combo, or just T5 on its own? I also noticed that the CLIPTextEncode (with T5 + CLIP) comes with its own CFG guidance setting, and I’m a bit unsure how to configure that properly.
I’ve always used it with both, in comfyui I do not have a CFG setting for clip, are you using Forge?
Here's to hoping my education goes good with this model!
Can you provide an example of your negative prompt for the Pro_Alpha version? I want to compare the results with my own. Looking forward to Pro_Beta as well :)
I rarely use negative prompts. As opposed to stable diffusion, it is not mandatory. HOWEVER! When I generate something and I see something I do not like, I include it to my negative prompt (keywords only) and it magically disappears.
The Pro-Beta is here !! =) Next up, actual release, in a few months probably XD
Which low step Lora is more compatible for 4-8 steps?
What sampler and scheduler to use for the pro-beta? All I am getting are blurry images :(
DPM++ and beta, I often use an upscaler too (SD ultimate upscale with Pandora-Dev as the model). The Pro model is awesome at understanding your prompts and generating results you wouldn’t get with a regular flux-dev model; but once you get the image, the time it would take to upscale is just not worth it.
@C47HERINE Oh, I was on euler + beta like dev-beta. The pro-beta just gave me blurry messes. I'll try dpm++ and beta.