@doctordiffusion.bsky.social
Modern mad scientist latent space explorer.
#photoscanned a bunch of my father’s film slides from the #1970’s / #1980’s. He had a pretty good eye back in the day if I do say so myself. Maybe I could convince him to pick up a camera again 📸
Should make for an interesting #machinelearning dataset.
Should make for an interesting #machinelearning dataset.
March 7, 2025 at 2:05 PM
#photoscanned a bunch of my father’s film slides from the #1970’s / #1980’s. He had a pretty good eye back in the day if I do say so myself. Maybe I could convince him to pick up a camera again 📸
Should make for an interesting #machinelearning dataset.
Should make for an interesting #machinelearning dataset.
🎉 SD3.5 Lives! 🎨🤖
I’ve just released Absynth 2.0, my fine-tuned enhanced base model for Stable Diffusion 3.5 Large, and I couldn’t be prouder of the improvements it brings to SD3.5! 🌟
Check it out on my Civitai or Huggingface.
#AIArt #StableDiffusion #GenerativeAI #machinelearning #aiartcommunity
I’ve just released Absynth 2.0, my fine-tuned enhanced base model for Stable Diffusion 3.5 Large, and I couldn’t be prouder of the improvements it brings to SD3.5! 🌟
Check it out on my Civitai or Huggingface.
#AIArt #StableDiffusion #GenerativeAI #machinelearning #aiartcommunity
January 19, 2025 at 6:27 AM
🎉 SD3.5 Lives! 🎨🤖
I’ve just released Absynth 2.0, my fine-tuned enhanced base model for Stable Diffusion 3.5 Large, and I couldn’t be prouder of the improvements it brings to SD3.5! 🌟
Check it out on my Civitai or Huggingface.
#AIArt #StableDiffusion #GenerativeAI #machinelearning #aiartcommunity
I’ve just released Absynth 2.0, my fine-tuned enhanced base model for Stable Diffusion 3.5 Large, and I couldn’t be prouder of the improvements it brings to SD3.5! 🌟
Check it out on my Civitai or Huggingface.
#AIArt #StableDiffusion #GenerativeAI #machinelearning #aiartcommunity
I ’ve done it again! civitai.com/models/1085468 New Detail Enhancement slider LoRA to greatly improve render quality of Stable Diffusion 3.5! Working De-Distilled Flux version coming soon #stablediffusion #sd3.5 #civitai #aiart
January 10, 2025 at 1:29 PM
I ’ve done it again! civitai.com/models/1085468 New Detail Enhancement slider LoRA to greatly improve render quality of Stable Diffusion 3.5! Working De-Distilled Flux version coming soon #stablediffusion #sd3.5 #civitai #aiart
Coming soon: a new “detail slider” to greatly improve results and quality of Stable Diffusion 3.5.
The first image is showing base render with SD3.5 alone and all other images are me increasing the strength of my newst tool.
Article on how I did it also in the works.
The first image is showing base render with SD3.5 alone and all other images are me increasing the strength of my newst tool.
Article on how I did it also in the works.
January 9, 2025 at 3:22 PM
Coming soon: a new “detail slider” to greatly improve results and quality of Stable Diffusion 3.5.
The first image is showing base render with SD3.5 alone and all other images are me increasing the strength of my newst tool.
Article on how I did it also in the works.
The first image is showing base render with SD3.5 alone and all other images are me increasing the strength of my newst tool.
Article on how I did it also in the works.
🚨 Big news in AI training stability! Massive strides across multiple architectures, including Flux Dev & Stable Diffusion 3.5. Before/After properly aligning the text encoders. Groundbreaking improvements! More details soon. 🔥 #AI #DeepLearning #StableDiffusion #FluxDev #AIRevolution #opensource
January 1, 2025 at 10:51 AM
🚨 Big news in AI training stability! Massive strides across multiple architectures, including Flux Dev & Stable Diffusion 3.5. Before/After properly aligning the text encoders. Groundbreaking improvements! More details soon. 🔥 #AI #DeepLearning #StableDiffusion #FluxDev #AIRevolution #opensource
A holiday greetings from AI me!
My latest LoRA of myself I trained with diffusion-pipe for HunyuanVideo.
My latest LoRA of myself I trained with diffusion-pipe for HunyuanVideo.
December 25, 2024 at 7:27 PM
A holiday greetings from AI me!
My latest LoRA of myself I trained with diffusion-pipe for HunyuanVideo.
My latest LoRA of myself I trained with diffusion-pipe for HunyuanVideo.
source.plus is a great resource for gathering images for fine-tune or LoRA training. Looking forward to using their Public Diffusion model trained on only public domain and CC0 images.
Source.Plus
Search, curate, and enrich media collections for AI training using the Source.Plus marketplace. Safe, consenting, high-quality training data. Public domain datasets for model fine-tuning.
source.plus
December 11, 2024 at 4:00 PM
source.plus is a great resource for gathering images for fine-tune or LoRA training. Looking forward to using their Public Diffusion model trained on only public domain and CC0 images.
Trained new Disco Diffusion LoRAs for Stable Diffusion XL, 3.5M, 3.5L & Flux Dev. So much faster than real disco. Get them here:
civitai.com/models/98518...
civitai.com/models/98518...
December 4, 2024 at 1:21 PM
Trained new Disco Diffusion LoRAs for Stable Diffusion XL, 3.5M, 3.5L & Flux Dev. So much faster than real disco. Get them here:
civitai.com/models/98518...
civitai.com/models/98518...
Disco Diffusion still hits different. I wish we could get it running as fast as modern ai image models.
November 26, 2024 at 9:44 PM
Disco Diffusion still hits different. I wish we could get it running as fast as modern ai image models.