- 105
- 44 038
FlowState
United States
เข้าร่วมเมื่อ 25 พ.ย. 2021
AI. Physics. Engineering. Tech.
There are Request Chats on Patreon and Discord for node requests or video requests for topics you want to see covered.
There's also an Image Chat on Discord for sharing your images. We have a weekly contest to decide the best images, and I pull the winners to feature in my videos every week. So go put your images there & vote!
Everything is open & free but if you like the work & wish to see further development, please consider a GitHub Sponsorship 🐙 or Patreon Membership ❤️ or one-time donations through CashApp 💵, Venmo 🐐 or Stripe 🪙
Also, please consider supporting our FlowState School on KickStarter and/or GoFundMe!
All links in channel bio!
Thank you so much to all the supporters! I promise to make you contributions worthwhile with more amazing nodes & high quality video content!
There are Request Chats on Patreon and Discord for node requests or video requests for topics you want to see covered.
There's also an Image Chat on Discord for sharing your images. We have a weekly contest to decide the best images, and I pull the winners to feature in my videos every week. So go put your images there & vote!
Everything is open & free but if you like the work & wish to see further development, please consider a GitHub Sponsorship 🐙 or Patreon Membership ❤️ or one-time donations through CashApp 💵, Venmo 🐐 or Stripe 🪙
Also, please consider supporting our FlowState School on KickStarter and/or GoFundMe!
All links in channel bio!
Thank you so much to all the supporters! I promise to make you contributions worthwhile with more amazing nodes & high quality video content!
Deep Dive on New Nodes & Workflow Hookup!
I had a really great conversation with a viewer on the stream yesterday and I think it will be VERY usefult for a lot of you. So I clipped it for those who don't want to sit through an 11 hr stream lol!
For requests, we have a Request Chat on Patreon & Discord, to ensure your request is seen and not buried in the TH-cam comments. It can be node requests or video requests for topics you want to see covered.
There is also an Image Chat on Patreon & Discord for sharing images and workflows. We have a weekly contest to decide on the best images, and I pull the winners to feature in my videos every week. So go put your images there and vote!
Please like, subscribe & share the video if you find it useful. It really helps me out a lot!
Patreon: www.patreon.com/flowstateeng/chats
Discord: discord.gg/KgxgXBjMX6
# Sponsorship
--------------------------------------------
Everything is open and free, but if you like my work and wish to see updates and new features please consider sponsoring my projects.
- FlowState Unified Sampler: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateUnifiedSampler.py
- FlowState Unified Model Loader: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateUnifiedModelLoader.py
- FlowState LLM Prompt & LLM Output: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateLLMPrompt.py
- FlowState Latent Chooser: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateLatentChooser.py
- FlowState Flux Video Diffusion: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateFVDSampler.py
The only way to keep the code and workflows open and free is by sponsoring its development. The more sponsorships, the more time I can dedicate to open source projects for everyone!
Please consider a Github Sponsorship 🐙 (github.com/sponsors/flowstateeng) or Patreon Membership ❤️ (www.patreon.com/flowstateeng/membership ), or one-time donations through CashApp 💵 (cash.app/$FlowStateEngineering), Venmo 🐐 (account.venmo.com/u/flowstateeng) or Stripe 🪙 (donate.stripe.com/9AQ3fm04OfqCgiAcMN).
I hope to have TH-cam channel memberships and support options unlocked very soon, and I will keep you posted on that!
Thank you so much to all the supporters, it is so incredibly humbling and validating to see people value my work enough to support its continuation! I promise to make you contributions worthwhile with more amazing custom nodes, and high quality video guides and demonstrations!
### Follow
--------------------------------------------
Patreon - patreon.com/flowstateeng
GitHub - github.com/flowstateeng
TH-cam - youtube.com/@flowstateeng
X - x.com/flowstateeng
Twitch - twitch.tv/flowstateeng
Rumble - rumble.com/user/flowstateeng
IG - flowstateeng
### Contact
--------------------------------------------
Email - info@flowstateengineering.com
### Corporate Sponsors & Affiliates
--------------------------------------------
Digital Ocean - m.do.co/c/a4ae3b3147ac ($200 CREDIT!!!)
VidIQ = vidiq.com/flowstateeng
### CREDITS
--------------------------------------------
Music by This Modern Day: FlowState - Life in Flux
TAGS
---------------------
ai art, artificial intelligence, machine learning, deep learning, ai tools, ai research, openai, open ai, anthropic, claude, google ai, stable diffusion, flux, black forest labs, open source ai, open source, opensource, llms, llm, mistral, llamma, llava, sora, ai news, ai, chatgpt, computer science, gaming PC, computer hardware, controlnet, controlnets, control net, control nets, comfy, comfyui, comfy ui, comfy ui flux, comfyui flux, comfyui tutorial, comfyui testing, testing
For requests, we have a Request Chat on Patreon & Discord, to ensure your request is seen and not buried in the TH-cam comments. It can be node requests or video requests for topics you want to see covered.
There is also an Image Chat on Patreon & Discord for sharing images and workflows. We have a weekly contest to decide on the best images, and I pull the winners to feature in my videos every week. So go put your images there and vote!
Please like, subscribe & share the video if you find it useful. It really helps me out a lot!
Patreon: www.patreon.com/flowstateeng/chats
Discord: discord.gg/KgxgXBjMX6
# Sponsorship
--------------------------------------------
Everything is open and free, but if you like my work and wish to see updates and new features please consider sponsoring my projects.
- FlowState Unified Sampler: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateUnifiedSampler.py
- FlowState Unified Model Loader: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateUnifiedModelLoader.py
- FlowState LLM Prompt & LLM Output: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateLLMPrompt.py
- FlowState Latent Chooser: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateLatentChooser.py
- FlowState Flux Video Diffusion: github.com/flowstateeng/FlowState-Nodes/blob/main/FlowStateFVDSampler.py
The only way to keep the code and workflows open and free is by sponsoring its development. The more sponsorships, the more time I can dedicate to open source projects for everyone!
Please consider a Github Sponsorship 🐙 (github.com/sponsors/flowstateeng) or Patreon Membership ❤️ (www.patreon.com/flowstateeng/membership ), or one-time donations through CashApp 💵 (cash.app/$FlowStateEngineering), Venmo 🐐 (account.venmo.com/u/flowstateeng) or Stripe 🪙 (donate.stripe.com/9AQ3fm04OfqCgiAcMN).
I hope to have TH-cam channel memberships and support options unlocked very soon, and I will keep you posted on that!
Thank you so much to all the supporters, it is so incredibly humbling and validating to see people value my work enough to support its continuation! I promise to make you contributions worthwhile with more amazing custom nodes, and high quality video guides and demonstrations!
### Follow
--------------------------------------------
Patreon - patreon.com/flowstateeng
GitHub - github.com/flowstateeng
TH-cam - youtube.com/@flowstateeng
X - x.com/flowstateeng
Twitch - twitch.tv/flowstateeng
Rumble - rumble.com/user/flowstateeng
IG - flowstateeng
### Contact
--------------------------------------------
Email - info@flowstateengineering.com
### Corporate Sponsors & Affiliates
--------------------------------------------
Digital Ocean - m.do.co/c/a4ae3b3147ac ($200 CREDIT!!!)
VidIQ = vidiq.com/flowstateeng
### CREDITS
--------------------------------------------
Music by This Modern Day: FlowState - Life in Flux
TAGS
---------------------
ai art, artificial intelligence, machine learning, deep learning, ai tools, ai research, openai, open ai, anthropic, claude, google ai, stable diffusion, flux, black forest labs, open source ai, open source, opensource, llms, llm, mistral, llamma, llava, sora, ai news, ai, chatgpt, computer science, gaming PC, computer hardware, controlnet, controlnets, control net, control nets, comfy, comfyui, comfy ui, comfy ui flux, comfyui flux, comfyui tutorial, comfyui testing, testing
มุมมอง: 360
วีดีโอ
Flux Lite vs. Flux NF4 - Which wins on Speed & Quality?
มุมมอง 52028 วันที่ผ่านมา
A new distilled Flux model from Freepik is here. Let's see how it compares to the Dev model and the NF4 model both in terms of quality and performance. For requests, we have a Request Chat on Patreon & Discord, to ensure your request is seen and not buried in the TH-cam comments. It can be node requests or video requests for topics you want to see covered. There is also an Image Chat on Patreon...
Chaining Flux LoRAs WITH EASE!
มุมมอง 66428 วันที่ผ่านมา
The new Unified Styler allows you to load & chain LoRAs with ease! Check if out at: github.com/flowstateeng/FlowState-Nodes For requests, we have a Request Chat on Patreon & Discord, to ensure your request is seen and not buried in the TH-cam comments. It can be node requests or video requests for topics you want to see covered. There is also an Image Chat on Patreon & Discord for sharing image...
Broken Glass Effect in ComfyUI with Flux & Stable Diffusion!
มุมมอง 301หลายเดือนก่อน
Broken Glass Effect in ComfyUI with Flux & Stable Diffusion!
Stable Diffusion 3.5 is HERE - Use with FlowState Nodes for MAX Effect!
มุมมอง 682หลายเดือนก่อน
Stable Diffusion 3.5 is HERE - Use with FlowState Nodes for MAX Effect!
Unified Model Loader - Super Convenient ComfyUI Node!
มุมมอง 507หลายเดือนก่อน
Unified Model Loader - Super Convenient ComfyUI Node!
Flux Video Diffusion & New Node Updates!
มุมมอง 1.6Kหลายเดือนก่อน
Flux Video Diffusion & New Node Updates!
Flux + SVD - Joined in Harmonious Holy Matrimony!
มุมมอง 1.4Kหลายเดือนก่อน
Flux SVD - Joined in Harmonious Holy Matrimony!
HUGE UPDATES! The BEST ComfyUI Custom Nodes You’ll Find!
มุมมอง 1.9Kหลายเดือนก่อน
HUGE UPDATES! The BEST ComfyUI Custom Nodes You’ll Find!
Create Your Own Custom Nodes for UNLIMITED Functionality!
มุมมอง 847หลายเดือนก่อน
Create Your Own Custom Nodes for UNLIMITED Functionality!
FlowState Custom Nodes - Get OCD Level Functionality Instantly!
มุมมอง 604หลายเดือนก่อน
FlowState Custom Nodes - Get OCD Level Functionality Instantly!
Master Flux Control Nets for ULTIMATE Precision Control!
มุมมอง 1.8Kหลายเดือนก่อน
Master Flux Control Nets for ULTIMATE Precision Control!
Flux IP Adapters For Incredible Style Transfer!
มุมมอง 2.8Kหลายเดือนก่อน
Flux IP Adapters For Incredible Style Transfer!
Styling Images with Flux LoRAs for Ultimate Aesthetic Impact!
มุมมอง 321หลายเดือนก่อน
Styling Images with Flux LoRAs for Ultimate Aesthetic Impact!
SERIOUSLY A GAME CHANGER - Flux Sampler Parameters!
มุมมอง 2Kหลายเดือนก่อน
SERIOUSLY A GAME CHANGER - Flux Sampler Parameters!
Boost Performance with NVIDIA Control Panel Settings for Flux!
มุมมอง 599หลายเดือนก่อน
Boost Performance with NVIDIA Control Panel Settings for Flux!
Powerful Results with Flux Control Nets & IP Adapters! (Plus a Secret)
มุมมอง 1.4Kหลายเดือนก่อน
Powerful Results with Flux Control Nets & IP Adapters! (Plus a Secret)
Pay Attention! Boost Your Image Quality with Clip Attention Multiply!
มุมมอง 450หลายเดือนก่อน
Pay Attention! Boost Your Image Quality with Clip Attention Multiply!
Achieve Inner Peace with This Squeaky Clean Flux Workflow
มุมมอง 300หลายเดือนก่อน
Achieve Inner Peace with This Squeaky Clean Flux Workflow
Master Any Skill with Rapid Prototyping
มุมมอง 231หลายเดือนก่อน
Master Any Skill with Rapid Prototyping
Ultimate Guide to Building an AI Ready PC from Scratch
มุมมอง 199หลายเดือนก่อน
Ultimate Guide to Building an AI Ready PC from Scratch
Turn Your PC into an AI Video Production Studio
มุมมอง 1.1K2 หลายเดือนก่อน
Turn Your PC into an AI Video Production Studio
Searge LLM in ComfyUI is the SECRET to Prompting LIKE A BOSS
มุมมอง 9372 หลายเดือนก่อน
Searge LLM in ComfyUI is the SECRET to Prompting LIKE A BOSS
VIT-14 Revolutionized My Text Results Instantly!
มุมมอง 5222 หลายเดือนก่อน
VIT-14 Revolutionized My Text Results Instantly!
Install ComfyUI on External Drive to Save Space - Comfy Life Hack!
มุมมอง 2642 หลายเดือนก่อน
Install ComfyUI on External Drive to Save Space - Comfy Life Hack!
Save Time & Stay Organized with ComfyUI Manager (Sept. 2024)
มุมมอง 1322 หลายเดือนก่อน
Save Time & Stay Organized with ComfyUI Manager (Sept. 2024)
Create AI MASTERPIECES on ANY Gaming PC - Flux Dev & NF4! (Deep Dive 2024)
มุมมอง 3082 หลายเดือนก่อน
Create AI MASTERPIECES on ANY Gaming PC - Flux Dev & NF4! (Deep Dive 2024)
Create AI MASTERPIECES on ANY Gaming PC - Flux Dev & NF4! (2024)
มุมมอง 1792 หลายเดือนก่อน
Create AI MASTERPIECES on ANY Gaming PC - Flux Dev & NF4! (2024)
Create Stunning AI Art on ANY Gaming PC with ComfyUI! - DEEP DIVE Setup (2024)
มุมมอง 1502 หลายเดือนก่อน
Create Stunning AI Art on ANY Gaming PC with ComfyUI! - DEEP DIVE Setup (2024)
Very inconvenient interface. How can a Forge user find out that for GGUF you need to download nodes, how do you know that GGUF models are put into UNET. You can scan the catalog and let the scripts do the work, bring in the models, and download what you need.
king shit man
cool ! i still prefer to write the LoRAs into the prompt like <lora:flux1980s:0.2> , then use LoratextExtractor to separate sending the text to the prompt and LoRAs to MultiLoraLoader . the drop down interfaces are fine when the total number of loras you have is small but when after a certain amount it becomes totally unusable . in this way can browse and organize your LoRA folder in file explorer and copy paste the names into the prompt , i use civitai helper to download the preview image . the other benefit is that you can do simple text operations to for example use a random LoRA from your folder , or random strength , or scale all the strengths at once even using a maximum strength to scale them relatively . like imagine how unwieldly these loaders are if you are using 20 LoRAs in one image , and have 2000 LoRAs to browse thru .
I wanna sit on it
🤣
He kinda fine asf
thank you lol
I feel so, so ashamed of my cable management. I swear I will fix it when my new cables come in LOL
Did I read the length of the video correctly? Like 343 HOURS?
It's been going for a while. Need to update the music on there. Got some decent new stuff
Nice work! Thank you! Question: Why no support for GGUF models on the FlowState Unified Model Loader?
thank you, gguf support will come in the next couple weeks. unfortunately ist not as straightforward to implement as UNET, checkpoint, or NF4 support. So we have those 3 for now but I do plan to add gguf as soon as i have the time to work it in there
Hey, new to this whole open source AI thing. Bought my first desktop with a real GPU (4080) a couple months ago. I've managed to get this workflow running, but am unable to figure out which folder to move my other flux models to. I could only get the new lite one, and a couple of stable diffusion models to come up?
your SD and Flux NF4 models should go in your ComfyUI/ComfyUI/models/checkpoints folder. Your Flux Dev, Flux Schnell, Flux Lite, and Flux Pixelwave, should go in your ComfyUI/ComfyUI/models/unet folder. Your GGUF LLM models should go in your ComfyUI/ComfyUI/models/llm_gguf folder, which you will need to make because it isnt a standard folder in Comfy. Make sure the GGUF LLM models are "instruct" type models. Like this: huggingface.co/meta-llama/Llama-3.2-1B-Instruct
Nice, i was hesitant yesterday to interrupt you during the stream to ask a question about your workflow not working for me neither. Gonna check this to see if it fixes my issues edit : still getting FlowStateUnifiedStyler.apply_style() missing 1 required positional argument: 'unload_loras' unfortunately. I also noted that prompt output isn't showing anything and startup nodes is reloading everything for each generation instead of keeping the unchanged models in vram which leads to unbearable generation time for a single image.
yeah make sure youre not using an old workflow. comfy doesnt always automatically updated the nodes in the workflow. so if you install the new version of the nodes and use an old workflow, it the nodes wont look like the new nodes. we are working on a standalone application thats entirely separate from comfy but for now gotta deal with their quirks. for example, if i added/removed an input or output, and you install the new nodes with that update, but you load an old workflow, the node wouldn't show that updated input/output in the workflow. so then things get misaligned and nothing works. i have a feeling thats whats going on in your situation
that's really useful to see, was trying to reverse engineer building from scratch but this is super clear now, thanks!
wait until you see the new ones lol. about to do a push in a minute. will announce on patreon and discord. theres only 1 noodle (basically) in the whole workflow
brostate
Thanks for the great explanation. I’ve been learning how to use ComfyUI gradually over the past six months. I guess it’s time to start building nodes!
thank you! well check out my recent videos on custom nodes. i have built up a small suite of them that should cover just about everything you need out of comfy. might save you a bit of time and effort lol. also be sure to check out the discord if you havent. discord.gg/flowstateeng
had ot low security to be able to install from your git hub, still doesnt work. No workflows are embedded in the PNG's wasted hours over several days trying to use this. Why are you not registering your node properly with ComfyManager so we can download a safe non risky version?
I'd like to generate images faster but with good quality - should I go for Flux NF4 or Flux Dev with the Lora Turbo Alpha 8 steps from Alimama?
id stick to NF4. i tried the turbo alpha and didnt get great results. that said, i have someone on the discord who is really smart and he swears the turbo alpha is the way to go. so its possible i havent dialed in how to use it yet. in any case, its not much faster than NF4, if at all and NF4 always gives me results that are nearly identical to the Dev model. so its a safe choice. be sure to check out the discord if you havent. discord.gg/flowstateeng
I read some comments saying that loras trained from flux dev are not compatible with the lite model....
hey, ive gotten good results with loras on the lite model but i also havent tried a ton. in general i prefer NF4, but since its not compatible with a lot of loras and control nets, i prefer the dev model for that. i think the lite model still needs some work
Nice Video
thanks for all your dope work! for some reason, after installing, rebooting Comfy, still have tons of red boxes and missing nodes. I dragged in your example 'horror rarri' banner image as test. Any ideas? Seldom had this problem with other nodes. (also did full Comfy Update)
thank you! yeah im posting a video in an hour ish that should clear things up.
Really nice introduction, was going to build a few nodes myself and this has really got me inspired.
Quick question: where did the LLM Prompt & LLM Output go? I'm getting a 404 error even with the link-it was super convenient to use.
yeah, after installing i get LLM / python error
sorry for the confusion. that node grew to be more than just an LLM node. There is so much else baked into it. So I renamed it to FlowState Unified Prompt. The output is FlowState Prompt Output. I try not to change node names but that one really needed it. Lmk if you have trouble finding it. Be sure to check out the discord to if you haven't. discord.gg/flowstateeng
WooW Dat cool work 😀
Very nice, one hell of an update to the FlowState!
mrflow dun did it again
This is really impressive, keep it up! I'd say this is one of the few node collections I'm actually looking forward to using. Really appreciate the brilliance from these nodes!
Thank you! That means a lot. A ton of work has gone into the project, so that's awesome to hear! I'm building fast, pushing updates almost daily. Basically a continuous running beta until its perfected. This allows me to get the tools in your hands so A) you can have them and B) you can lmk if you run into any issues. The Discord and the live streams are full of knowledgeable people if you need any help! discord.gg/KgxgXBjMX6
nice! would like to see these at lower guidance say 3.5 which might make it less "hyperreal"?
I will have another Control Net video coming soon that will blow your mind. Join the streams sometime and I'll take your requests for anything you want to try. And get in the discord if you need help, have suggestions and/or want to chat and share you images for the weekly contest. discord.gg/KgxgXBjMX6
th-cam.com/users/liveMloTXASpr5I?feature=share
Bad assery
Yo, straight up fell asleep!
Hi flow, great video but disagree about loras, they do make a big difference when used correctly, especially for were the checkpoint model cannot do a certain task, the lora will add that ability, not just styles. PS I don;t know if its still the case but nf4 wasn't compatible with loras, it had to be gguf, fp8 etc
I've come around on them lol
Amazing video !
Thank you Six!
no way freakbob made it. biggest life accomplishment
Thanks for picking one of my pieces, the challenge was a lot of fun. For anyone who is looking for an awesome place to hang out, make friends, come up with creative ideas, make awesome AI art and learn new cutting edge techniques I'd recommend joining and becoming a part of the community over on the cord. We have people from all different knowledge backgrounds, we have people who use all sort of different avenues of AI like he said. We also have photographers that have a vast understanding of different composition techniques and how to achieve those great film like looks.
True, that's another thing I didn't think to mention. There's several professional photographers and audio/video editors there as well. Good point. Thanks @ThisModernDay, great images!
Hell yeah!
where does the ipadapter file go into which comfyui folder - models/ipadapter ? it is not showing up in the node
you'll want to put that one in the models/xlabs/ipadapters. the folder is created when you install their nodes
Your title says New Comfy Desktop App is Here! Can you tell me where you can download it if that's the case? Otherwise, not sure what your goal is.
did you watch the video? they released a promo video for the app. my video is just reviewing their promo and breaking down the new features we can expect to see. i do have an advance copy and will be posting a video of it today or tomorrow.
@@flowstateeng I get that but you said it was here now? I have the latest code release and can turn these settings on in my local host version but I don't see a download for the desktop all in one is what I was wondering?
yeah they have a waitlist to join for now unfortunately. its announced but i guess its technically not "here" yet. sorry for the confusion
@@flowstateeng No worries, I am on the waitlist, I just thought maybe you saw something I didn't! :)
Saving the workflows does not work at all as expected :) I just export them everytime just to make sure. If they solved this in the new version it would be greate. Also the modeldownloader will save some time.
yeah same here, ongoing bug. it used to work. i have an advance copy of the new desktop app and same bug there too unfortunately. make sure to get in the discord too! we got a nice community over there to help with bugs, questions, share image/vid generation techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6
Nice Video
thanks man!
Will ot finally work seamlessly with InsightFace?
that i do not know. can you hop in the discord and ask in the general chat? we got a nice community over there to help with bugs, questions, share techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6
This is already implemented in the Pinokio version. Check that out if you don't want to wait.
i have this error FluxSamplerParams+ 'ForgeParams4bit' object has no attribute 'quant_storage' I make sure I followed the instructions more than once
can you hop in the discord and ask in the bug chat! we got a nice community over there to help with bugs, questions, share techniques, etc. makes it easier to share screenshots of your CMD output, etc. discord.gg/KgxgXBjMX6
@@flowstateeng great community 👌
Thanks for these nodes! I might have missed this, but what does the latent multiply setting on the sampler do?
yeah no worries! make sure to get in the discord! we got a nice community over there to help with bugs, questions, share techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6 basically latent multiply gives your colors a little bump. i usually keep it around 1.15 - 1.25
Hi, thanks for this great work! By the way, if we use sdxl models how to change the cfg scale?
no worries! the "guidance" is the same as cfg on the sampler. make sure to get in the discord too! we got a nice community over there to help with bugs, questions, share techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6
I still can’t get your nodes to load. I’ve tried installation manually or through the comfy interface and neither works for me.
can you jump into the discord and post your screenshots in the bug chat channel? you can do patreon too if you dont have discord. discord.gg/KgxgXBjMX6 patreon.com/flowstateeng
Would like a LoRA loader too in the workflow
yep, thats next on the agenda. you should join the stream tonight. we're going to get started on that. make sure to get in the discord too! we got a nice community over there to help with bugs, questions, share techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6
Another suggestion for flowstate latent chooser: add a direct output for image so you don't have to add a vae decode in case you want to use an image comparator, for example
i think it does already no? if not ill change it. i had it that way at one point. good catch, thanks
Very interesting. For the next update you could add GGUF.
yep we are already planning that. several people have asked and i think its a good feature to have. make sure to get in the discord! we got a nice community over there to help with bugs, questions, share techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6
Thanks for this! I wonder if it's better to install it this way or thru the Pinokio app? I did it thru Pinokio and the only difference I see is that I don't have the Web Search option on mine.
Scratch that, I got web search working now as well.
awesome, glad you got it working. make sure to get in the discord! we got a nice community over there to help with bugs, questions, share techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6
One Loader to rule them all! Nice thanks.
you bet! make sure to get in the discord! we got a nice community over there to help with bugs, questions, share techniques, etc. its coming along nicely! discord.gg/KgxgXBjMX6
You in Pensacola too?
Yessir
Near Palafox & Burgess area