Get the TOX here >>> www.patreon.com/posts/104339390
New 'ControlNet' parameter page
ControlNet active (Usecontrolnet) must be On when starting stream, but can be toggled when streaming to control if used during generation.
Weight parameter (Cnweight) will update the strength of controlnet when streaming 2024-05-16 04:49:31 +0000 UTC View Post
Changelog / announcement here >>> https://www.patreon.com/posts/104339324/
2024-05-16 04:48:25 +0000 UTC View PostUpgrade Guide
Similar to previous version, just copy and paste your Basefolder parameter ( and depending on your setup change other settings in the operator) and it should load your previous working model list. Once you select a Main SD Model, you should be ready to go.
Only a few changes but exciting ones :
Mac (Apple Silicon) supported ! (Will need capable apple silicon ! slower than nvidia cards but working )
Model Preset Config system added - allows for all Models to be set with a preset... (handy for recalling tensorrt setups)
Big shoutout / thank you to @barretto for the help getting this working cross platform.
For mac, if you change resolution or restart stream, it might have trouble finding the buff...
2024-05-03 19:36:14 +0000 UTC View Post[ EDIT - 0.1.9 Version fixes possible error when switching resolutions + issues with finding shared memory ]
Changelog / announcement - www.patreon.com/posts/streamdiffusiont-102788281
Upgrade Guide
2024-04-21 22:22:54 +0000 UTC View Post
Similar to previous version, just copy and paste your Basefolder parameter ( and depending on your setup change other settings in the operator) and it should load yo...
Get the TOX - www.patreon.com/posts/102788463
Important ! This version of StreamDiffusionTD replaces NDI + Spout with a faster solution for improved performance. However, I am planning to reintegrate ndi in the next version in a way that makes it easier to run non locally. 0.1.8 will only work locally on a single machine, but will be expanding support for non local.
[ EDIT - 0.1.9...
2024-04-21 22:22:39 +0000 UTC View PostMinor update but should help if you were having trouble with importing OpenAI library. Thanks to all that provide helpful feedback / report bugs.
(My OpenAI library install process is Windows Only for now. Mac support on stuff coming soon. If you 'pip install openai' with anaconda or tdpip or any other way to import that library, it will work as well.)
Fixes + Changes
Improved 'OpenAI Import RESET' Pulse in Config for when setting up the operator.
Adde...
2024-04-21 03:01:19 +0000 UTC View PostGeneral Improvement + bug fixes on Text2Speech and Speech2Text API tool.
See more about the tox - https://www.patreon.com/posts/92826648
Prevents Audio playing last TTS output when project opens.
Now has separate Timers for transcript/speech
Improved Parameter layout + clarity
Removed some confusing / vague errors
If i...
2024-04-21 02:16:57 +0000 UTC View Post
hey all ! Stability announced that SD3 is available via API only for now (model weights hopefully coming soon) ! But if you wanted to try it out in TD before that... here is TOX for testing out / generating with SD3 + SD3-Turbo models via Stable Diffusion 3 API. It should only require a stability API key (get API Key here).
all API parameters exposed (-1 and 0 are both random Seed)
works with both t...
2024-04-18 21:34:49 +0000 UTC View PostImproved Model Support !
SDXL-Turbo supported !! ( but no TensorRT yet ... working on that )
stabilityai/sdxl-turbo in Main SD Model ID/Path... see more
SDXS, a recently released 2.1 based model is now supported. faster than sd-turbo, but not good for img2img. very smooth for text2image interpolation though. SDXS + TensorRT beats sd-...
2024-04-11 22:18:59 +0000 UTC View PostImportant ! There is no need to reinstall the repository if you have a working previous version. To upgrade to 0.1.7, simply copy the Basefolder path from a previous working version and paste it into the Basefolder parameter of the new operator. Adjust the necessary parameters, models, and settings accordingly.
changelog here
Hey ! Last month I created a forum called #tools-tips-toxes in the discord server that I will be posting some hopefully helpful stuff and also a few others have posted sharing some tools / techniques. Don't miss out on all the awesome stuff ! More coming soon.
( I will post my tools + toxes here as well, but there is definitely a lot of info / knowledge / tips / tools in the discord if you haven't joined. Link to the discord is in the 'About' parameter page of most recent TD ops !! ) 2024-04-09 05:35:26 +0000 UTC View Post
TOX + TOE included (also available via discord)
https://discord.com/channels/1026335703598104576/1227114693210931291
2024-04-09 05:33:26 +0000 UTC View Post
Hey ! Put together this video to introduce and cover the basics of ChatTD, which I updated last week. Excited to show the UI I have for the operator and very interested to hear how / what you build with it ! The conversation management also allows for sharing of chats and easier iteration of LLM integration within TD networks. See the release notes / more info here !
The operator is availa...
changelog / features - https://www.patreon.com/posts/chattd-2-0-0-big-100989682
2024-03-24 19:52:25 +0000 UTC View PostI have been wanting to make UI for this operator and build out more features with the image models to allow for better template building and conversation sharing. This one has a ton of new tricks I've found for UI and tool building. Available for dotexplorer tier for now! Interested to see what/how you build + experiment + use the operator for ! 🔥
More info / features:
Multi-API Support ! 🤹
- over 120 models available in total !
- 2024-03-24 19:50:50 +0000 UTC
View Post
Can't wait to see what you make with it ! For feedback / troubleshooting, discord is best !
2024-03-06 00:39:23 +0000 UTC View Post
Version 0.1.0 is a much improved version with more information visible about the workflows, making it easier to quickly load in workflows and iterate within TouchDesigner. ComfyTD still follows the same logic / flow as the ComfyTD Deep Dive video I posted last month, but I will be sharing more workflows / videos to expand and show some more specific setups soo.
TOX is available in a separate post !
Major New Features:
Solid update with a few more additions for working with models and a few bug fixes on installation !
If it was working for you before, there is no need to reinstall/upgrade the streamdiffusion repo. Load in the new tox and switch out the Basefolder, models, etc and you are all set.
Changes:
Installation Guide ! New 'View Instructions' pulse in Install page will open helpful instructions !!! check it out !!
'Lora' parameter page is renamed to 'Models' + more i...
2024-02-23 04:54:33 +0000 UTC View Post( 0.1.4 ) Fixes config / model error which would happen when first loading 0.1.3 due to not removing items from config.
Apologies if that caused any confusion, but glad it was pointed out to me so I could catch after only a few hours. Thank you to all who report bugs !!
Operator has been updated in the 0.1.3 post.
2024-02-15 05:15:28 +0000 UTC View PostTxt2img, offline support, working models menu and more!
( 0.1.4 ) Fixes Config error which might happen when first loading 0.1.3 due to not removing my custom VAE path.
Txt2img now available ! 'Sdmode' menu added in Settings 2 page
- This can be updated live when Streaming.
- If Sdmode is set to txt2img, then the stream will not be effected by the input image
- All settings still have an effect on the output, including the step sliders.
Allows usage of external libraries inside of TouchDesigner + managing them with a venv. This is my current preferred method and have been using a lot more recently after packaging into a pretty functional tool. Updated fully for 2023 and python 3.11 but also supports using any version of python installed on the computer.
Full extension class as well ! README included for more info.
Mac compatibility coming soon.
Demo network with a few different approaches on prompt blending with a few notes explaining each.
+ lots of chops/dat tricks throughout
Works for seeds too ! ( Seed table included )
Intended to be explored from the Top of the network down.
2024-02-12 04:40:44 +0000 UTC View PostPrompt Blocks added for mixing multiple prompts with per prompt weights
- 'Normalize Prompt Weights' toggle parameter
- 'Total Weight Slider' parameter to set amount to normalize prompt weights.
Seeds Blocks added for mixing multiple seeds with per seed weights
- 'Total Seed Noise Multiplier' parameter to boost/reduce overall noise
- NOTE: Mixing the same seed isn't recommended, but mixing different seeds should blend.
Improved...
2024-02-06 03:14:15 +0000 UTC View Post
Just finished putting together an attempt at a full explainer for ComfyTD ! I hope you enjoy !
2024-01-29 04:53:59 +0000 UTC View PostEvery parameter has help text ! Check parameter help text (alt + mouse rollover) if any parameter is confusing / unclear !
Changes + Additions:
- Breaking change ( only effects advanced setups / extension class manipulation ) - Parameters that began with 'Cui' now start with 'Ui'
- Added kwargs for the Generate() function that will take any of the API Parameters as optional inputs to replace the parameter value. example: op('ComfyTD').Generate(Seed=123456, Steps=20, Text...
Important Fix for NDI that prevents memory build up
( at the time of this post, I also uploaded v0.1.1 as the operator in the 0.1.0 TOX post )
[ Updated attached TOX to 0.1.1 - Important Fix for NDI that prevents memory build up ]
- Automated setup for StreamDiffusion.
- Stream launch and control from TouchDesigner.
- Live parameter updates for prompt, step schedule, seed, guidance, and delta.
- Support for multiple LoRA models.
- Stream performance monitoring (FPS).
- Integrated NDI and Spout ops.
- Callback functions: onReceiveFrame, onStreamStart, onStreamEnd.
- CHOP outputs for stream status, f...
✨ Hey friends! I'm excited to share this install/start-up guide for StreamDiffusionTD.
( TOX available now in separate post )
This operator is built in TouchDesigner 2023 and won't be compatible with earlier versions! Also, projects saved in 2023 cannot be opened in 2022. See the release post for more - Touchdesigner 2023 Official Update
🛠 Installation Tips!! (working on a readme + smoother op install process)
Python v 3.10.9 💎is tested and working! Other versions w...
✅ TD 2023 Version Compatibility: This TOX was created in 2022, but works in 2023 as well. Please note that projects saved in 2023 cannot be opened in 2022. See the release post for more - Touchdesigner 2023 Official Update
🛠️ 'Connection' Parameter Fixed: No longer will always display as 'On'.
❌ Fixed Red X Issue on Load 2023-12-14 05:51:23 +0000 UTC View Post
CAUTION! This is currently only tested on 2022. I will likely need to test / update this operator for the 2023 version.... but happy TD release ! 🎉
New / Improved:
📜 Improved Parameter Menu Parsing: Enhanced parsing of parameter menus using ComfyUI node data...menu items / parameter ranges will now be what they are in ComfyUI.
🏌️♂️ Update Range Toggle in Display Settings: The new toggle feature in display settings prevents interface lockout with each generation, imp...