AI as Rendering Engine: Ambrosinus Toolkit v1.2.1
Вставка
- Опубліковано 12 вер 2024
- 📣 Since Ambrosinus Toolkit v1.1.9 is possible running Stable Diffusion and the wonderful neural network named ControlNET locally via Grasshopper.
New components are based on the wonderful project of AUTOMATIC1111 which exploits all the power of Stable Diffusion entirely locally. But it doesn't end here, in fact through the expansion that integrates a second neural network, ControlNET v1.0 (developed by Stanford researchers) it is possible to use all the power of AI in the main creative activities carried out by architects and designers - AI as a rendering engine! (or something really promising 🧐).
💡With v1.2.1 is possible to interact with the webui-user.bat file, with Rhino views (so pass your model screenshot directly in the AI engine to get a conceptual render) and Upscale the image!
✨ The Ambrosinus-Toolkit project is always under development.
Stay tuned!
#AUTOMATIC1111 #ControlNET #AmbrosinusToolkit #DEV #Grasshopper #AI #AIasRenderingEngine #C #coding #rhino #lucianoambrosini
🧾 Files + full description - bit.ly/SDandCN...
👉 About me - bit.ly/AboutMe...
⏱️ Highlights
00:07 Main settings
00:57 Set the Dataset model checkpoint (for instance mdjrny-v4)
1:24 Open a Rhino file
2:00 ViewCapture component
3:27 If you encounter an issue when running the CMD process?
4:11 View 01 rendering
5:05 View 02 rendering
5:29 Image Upscaling
6:09 OpenDir component (wip)
6:32 Upscaled image embedded info
Can't believe I'm the first person to comment, but for all your amazing work, thank you! You are truly legendary! And thank you for brining such incredible technology to Rhino/Grasshopper. I hope this projects' trajectory is sky-high and wish you a huge adoption from the community.
Thanks so much for your words! I really appreciated that. Free projects often have this effect and require more time and word of mouth to be truly appreciated... But feedback like yours gives me a boost for my "experiments in the garage" 😉 thanks!!
Great video and update! Thanks for your effort and time, I really appreciate that you are sharing these with us!
Great video, Thanks! Have u faced same problem as me? Text to image plugin renders perfectly but I'm having problems when it comes to existing pictures. Everything works correctly in the browser
Thank you for your message. Honestly I have not encountered this problem yet...but I will test it again asap. Thank you for your feedback
Hi! I've followed all steps but I am getting stuck at step number 2, I dont know where to download SD checkpoints :(
Hello, and thanks for your message! Ok well-done, you can find many Stable Diffusion models (checkpoints) on civitai website. After downloading whatever you prefer you must put that model into this folder: "\\stable-diffusion-webui\models\Stable-diffusion\". Hope this will be useful :)
Great tool man, i~m having some problem with the preprocessor resolution being low when SD activatede from rhino, how could i increase this resolution
Hi Gabriel thank you for your feedback!
Do you mean WxH size? Did you change them? Have a look also at the post-process Upscaling of the image with the dedicated component.
Anyway, if you have more specific questions and you want to share your gh file or screenshots please feel free to reach out to this discourse forum and please, have a look also here: tinyurl.com/GHAIeng-upd2
Hello! The LA_SDopts_loc tells me: ' Remote server error: (404) Not Found'
Any leads on that? Thanks!
Hello! Thanks for your feedback. I have checked the file but seems everything is ok. I have run this file SD_Local_ControlNET_v10-11_v124_base&server_demo-01
Do you have more info to share with me by email? Please, let me know. 🙂
@@LucianoAmbrosini I'm figured out how to solve it thanks! :) To install CN 1.0 instead of 1.1.x from the extension tab, do I need to do something special?
@@georgeneverland896 Cool! You should be able to use both versions, for example, I have two separate folders in one I installed A1111 and CN1.0 and in the other A1111 and CN1.1.X. Generally, some errors are related to the installation procedure of Python libraries (Python included) ...but I have made some videos, I hope they can be useful to you, please take a look here: ambrosinus.altervista.org/blog/ai-as-rendering-eng-sd-controlnet-locally/
@LucianoAmbrosini thank you! My question is even simpler: Automatic1111 automatically installs the latest version published on the main branch, so is it enough for me to download the 1.0 models linked in your text guide?
@@georgeneverland896 Yes, the ControlNet extension could be v1.0 or 1.1.x depends on which link you pass in the box installation reachable from the WebUI>Extensions>Install from URL as described in my text guide. 👍
Hi Luciano, thanks for this tool! I just installed Automatic1111 and the Ambrosinus Toolkit 1.6. Now I am following your instructions but after going throught the step "Start" this message pops up:
Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug 1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)]
Version: v1.7.0
Commit hash: cf2772fab0af5573da775e7437e6acdca424f26e
Launching Web UI with arguments: --api --lowvram --xformers --precision full --no-half
Style database not found: C:\Users\xxxxxxxx\Desktop\SDlocal\stable-diffusion-webui\styles.csv
*** Error loading script: api.py
Traceback (most recent call last):
File "C:\Users\xxxxxxxx\Desktop\SDlocal\stable-diffusion-webui\modules\scripts.py", line 469, in load_scripts
script_module = script_loading.load_module(scriptfile.path)
File "C:\Users\javier.goca\Desktop\SDlocal\stable-diffusion-webui\modules\script_loading.py", line 10, in load_module
module_spec.loader.exec_module(module)
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "C:\Users\javier.goca\Desktop\SDlocal\stable-diffusion-webui\extensions\webui-controlnet-v1-archived\scripts\api.py", line 91, in
ControlNetTxt2ImgRequest = create_controlnet_request_model(StableDiffusionTxt2ImgProcessingAPI)
File "C:\Users\javier.goca\Desktop\SDlocal\stable-diffusion-webui\extensions\webui-controlnet-v1-archived\scripts\api.py", line 82, in create_controlnet_request_model
'controlnet_units': (List[ControlNetUnitRequest], Field(default=[], docs_default=[ControlNetUnitRequest()], description="ControlNet Processing Units")),
NameError: name 'List' is not defined
Do you know what can be the issue? Thanks in advance
Hello, thanks for your inquiry and interest in the toolkit! I suggest trying the demo file that you can find here: github.com/lucianoambrosini/Ambrosinus-Toolkit/tree/main/AI_components/SD_ControlNET_locally
that one with v1.2.5 in case of issues, please let me know about it by email or on the official forum where we have solved several topics 😊. I will wait for your feedback, thanks!