Generate images from text for free
Create your own incredible images from text with our free AI image generator. It's easy and fun!
2 years ago
Credits to @DBFISH for the prompt
Created 2 years ago Β· 2 commentsΒ· 0 likes
SDXL 1.0
Credits to @DBFISH for the prompt "Something's Vexing Her" - made with @NightCafeStudio
Creation Summary by Vincent
Renaissance Woman in Gown: Photorealistic Masterpiece
A photorealistic portrait of a Renaissance woman in a ball gown. The image has a dreamy, celestial quality with soft light and magical particles, reminiscent of a classical painting.
Created by Tgr22 on Oct 9, 2023 using the SDXL 1.0 AI image generator model.
Creation Settings
- Text Prompts
- & Renaissance Woman trying on a Ball gown Masterpiece portrait photorealistic HD Medium Shot striking hyper real painting Andrea Kowch Michelangelo Leonardo Da Vinci Jackson Pollock Titian light dust Post-Production SFX detailed intricate elegant realistic maximalist super detailed 8k resolution Substance ...Weight: 1.116K high res 4080x4080px maximalist , ζιΊΊ θΈθ‘, ζιΊΊ θΈθ‘, lady portrait...Intricate_Precision_Meticulous_Elaborate_Exquisite_Detailed_Complex_Fine-grained_Obsessive_Perfection_Microscopic_High-definition_Elaboration_Richness_Nuanced_princess_Ultra-realistic_Hyper-realism_princess_Fine-detailing_Thoroughness_Prolific_Comprehensive_Multilayered_Intensive_Microcosmic_Detailed-oriented_Masterful_In-depth_Exactitude_Dense_Profound ...Weight: 0.7Dysmelia:_A general term for limb_deformities,_which_can_include_hand_deformities. Polydactyly:_Having_extra_fingers_or_toes. Syndactyly:_Fusion or webbing of the fingers or toes. Brachydactyly:_Abnormally_short_fingers_or_toes. Clinodactyly:_Curvature or deviation of the fingers or toes. Arachnodactyly:_Abnormally ...Weight: -1.1opposite prompt:2.0) flatplaneplanets: porcelain lattice paper: straight lines contained and conservative gradients of smoothness. f1_false_theater,_dull_background._sub_480p,_low_res,_minimalistic,_false_theater,_sharp_edges,_low_poly,_mediocre,_abstraction_black, emotionless+dull,sad+dullWeight: -3
- Model
CKPT- Initial Resolution
Medium
- Aspect Ratio
3:4
- Runtime
Medium
- Overall Prompt Weight
70%
- Sampling method
K_DPMPP_2M
Well, this is Cool - Mostly I've been picking through your Prompts and they're fascinating. Unfortunately, Stable Diffusion (pick your version) is not able to deal with minutia. The Most Important part of a Prompt is the first paragraph - usually about 60 tokens (roughly 'words' of the First Prompt I don't know the metrics for how subsequent Prompts are incorporated - especially for SDXL1.0 which doesn't handle them brilliantly - I have had some good results and many bad ones. By the way the 'underline' "trick" is no longer necessary as Prompts will now tolerate more than the 60-75 tokens. When we can command Stable at the very granular level you are seeking to work at, it will be thrilling... By then SD will render in Realtime and we'll be talking to it. Thank you for evolving some of my work - I really do appreciate it. Remember the single most powerful tool we have for getting the most out of Prompts is the Start Image.