STABLE DIFFUSION AND STUFF NEWS : Chronological record of news

[March 12]

1- Blender : Character bones that look like Openpose for blender _ Ver_5 Depth+Canny Section 5.4
x- FAQ : the webui LoCon extension supposed to work with the addnetwork extension for Loha too? not for now
2- Extension deforum added Section 8.2
waiting section
3- Extension Removes background after each image generation : https://github.com/ilian6806/rembgr /// https://github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/8541
4- Extension Cutoff: Cutting Off Prompt Effect /// This is an extension for stable-diffusion-webui which limits the tokens' influence scope. https://github.com/hnmr293/sd-webui-cutoff

[March 11]

0- Attention: The news section from March 6 to March 12 will be move an another link. This link its not available yet. will be shown at the beginning of this section on the 13th, so that if you missed something new, you can go check it out, as well as keep a temporary record of the events that I have witnessed.Thanks you four time.
0.5- Any link referring to the creator of this rentry will be published on this site. Any link referring to this site, like the one I commented recently, will be added in this rentry.
1- Controlnet 2.x Scribble was released ! /// Controlnet model: https://huggingface.co/thibaud/controlnet-sd21 Section 4.1.1.2
2- Extension -"blip2" is a sd extension that generates image captions with blip2. Using that caption as a prompt may help you get closer to your ideal picture. /// https://github.com/Tps-F/sd-webui-blip2 Section 6.2
3- Extension "Blip2 captioner" - BLIP2 captioning tool as an extension of AUTOMATIC's WebUI. https://github.com/p1atdev/stable-diffusion-webui-blip2-captioner Section 6.2
4- Script to use locally: https://github.com/derrian-distro/LoRA_Easy_Training_Scripts , to train, extract and merge
5- Update NoCrypt Colab Section 3.1.2
6- Extension "tunnels": Tunneling extension for A1111. Cloudfare tunnels serves as a proxy between your internal LAN devices and the outside world without exposin your WAN IP adress - https://github.com/Bing-su/sd-webui-tunnels //// Section 6.7 To learn more, please visite https://youtu.be/ZvIdFs3M5ic
7- New sampler method "UniPC" A Unified Predictor-Corrector Framework for Fast Sampling of Diffusion Models was added to A1111 webui Section 10 : Papers
8- Article about LLMs: Large language models are having their Stable Diffusion moment // https://simonwillison.net/2023/Mar/11/llama/ //// Extra Burguer

[March 10]

1- A new day ! Hi everyone, I am pleased to announce that there is a guide with FAQs about Stable Diffusion, take a look. New FAQs will be added over time. Ideal for quickly answering questions. The author of this FAQ is ao汁. Congratulations on the incredible work! https://rentry.org/tohoaifaq
2- Possible improvements to DPM++ 2M Karras sampling: https://github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/8457
3- Paper- Cones: Concept Neurons in Diffusion Models for Customized Generation. https://arxiv.org/abs/2303.05125 /// Extra Burguer
4- Paper- GigaGAN: Scaling up GANs for Text-to-Image Synthesis Extra Burguer
5- Extension "Corridor Crawler Outpainting" - An automatic1111/stable-diffusion-webui extension for generating hallways with Stable Diffusion. It was designed to use depth-to-image model at 512 resolution, but it could be tweaked to work with other models: https://github.com/brick2face/corridor-crawler-outpainting Section 6.3
6- Extension "Bayesian Merger" ! - An opinionated take on stable-diffusion models-merging automatic-optimisation.Why not auto-merge extension? Brute force == long time to wait. With this method you can get away with number of runs ! : https://github.com/s1dlx/sd-webui-bayesian-merger 7.3.1.2.1.2
7- Extension "3D Model Loader" !! - Allows you to load your local 3D model/animation inside webui, then send screenshot to txt2img or img2img as your ControlNet's reference image: https://github.com/jtydhr88/sd-3dmodel-loader Section 6.4
8- Openpose editor: https://zhuyu1997.github.io/open-pose-editor/ //// Section 5.7.1
9- Controlnet 2.x Openpose model was released!! https://huggingface.co/thibaud/controlnet-sd21/tree/main //// Section 4.1.1.2
10- Update Stable Diffusion Interactive Graph Section 2.1 , please visit https://sdtools.org/ to interact . Good Work!
11- Controlnet 2.x Mod pruned !! .Safetensors version (pruned to only 730mb) are now available!: https://huggingface.co/thibaud/controlnet-sd21/tree/main //// Section 4.1.1.2
11- Niakonichan Twitter Account official available now: https://twitter.com/niakonichan8480 It helps us a lot if you mention me in something you see on Twitter
12- Weekly TT: Who Is Fantasy.ai? by Aitrepreneur: https://www.youtube.com/watch?v=_qcHRYpehUs
13- ATTENTION COUPLE: An improved version of Latent couple(or two shot, section) that generates multiple characters separately! https://note.com/gcem156/n/nb3d516e376d7 . Remember, if you see a japanese text, use translate if you not speak the lang. Demo Google Colab: https://colab.research.google.com/github/laksjdjf/sd_colabs/blob/main/Attention_couple.ipynb /// waiting for section

[March 9]

1- NEW !! Tutorial- How to use ComfyUI: https://comfyanonymous.github.io/ComfyUI_tutorial_vn/ Section 9
Work in progress
2- ELITE- Github: https://github.com/csyxwei/ELITE soon Section 2.15
3- The-Last-Ben Colab ISSUE - HOW TO SOLVE https://github.com/TheLastBen/fast-stable-diffusion/issues/1706
4- Colab Upgraded Python from 3.8 to 3.9
5- New Consistent VAE "Blessed": https://huggingface.co/NoCrypt/blessed_vae Section 4.1.2
6- LyCORIS Lora beYond Conventional methods, Other Rank adaptation Implementations for Stable diffusion. A project for implementing different algorithm to do parameter-efficient finetuning on stable diffusion or more. This project is started from LoCon. https://github.com/KohakuBlueleaf/LyCORIS LoCon is moved to LyCORIS project See Section 2.7
7- LoHas models start to emerge. Will this be the definitive evolution of LoRA? Section 2.7.2
8- Fast_PNG_Info - New extension to extract metadata more faster. It allows you to view prompt from an image without the need to upload it to a server, resulting in fast pnginfo. This is particularly noticeable when running the webui in a public URL, such as colab. https://github.com/NoCrypt/sd-fast-pnginfo Section 6.1.3
9- Spanish NoCrypt Colab available now! Section 3.1
10- Controlnet model HED for sd 2.1 is now released! Section 2.10.2
11- Extension - SD web UI bilingual localization: https://github.com/journey-ad/sd-webui-bilingual-localization Section 6.7
12- Finally after hours the issue about LastBen Colab was solved! Thanks to the developers.[March 8]

[March 8]
1- Controlnet model Canny 2.x == 2.10.2==
2- Online Resource- Helper to use Latent Couple (Two Shot) Section 5.8
3- "Sad Talker" Extension . A realistic and stylized talking head video generation method from a single image and audio Section 2.14
4- The arrangement order of prompts has a significant impact on the image output results (especially in Hires.Fix mode).
https://civitai.com/models/17049/the-impact-of-prompt-order-on-result-generation
5- Controlnet model depth 2.x :https://huggingface.co/thibaud/controlnet-sd21 2.10.2
6- 2 API to Controlnet extension: https://github.com/Mikubill/sd-webui-controlnet/wiki/API
7- Breadboard 0.4.5 Released. https://breadboard.me/ Section 5.3.1

[March 7]
1- Like a tunning script for vae Section 7.3.5
2- "X-Decoder"-Instructional Image Edition Section 2.13
3- T2I-ADAPTER "Transfer_Style_Model" == 4.1.1.1==
4- "Local Latent upscaLer(LLuL)"-An extension for sd-webui which lets you to upscale latents locally Section 6.3

[March 6]
1- "VAE Merger" A WebUI extension for merging VAE Section 7.3.5
2- "GLIGEN": Open-Set Grounded Text-to-Image Generation -- [2023.03.05] Gradio demo code is released Section 2.12
3- "Layer divider"-To convert generated images AI to .psd file (potoshop, clip studio, etc) Section 5.6