Good video, I feel like the audio track of your voice could have been a bit louder compared to the background music.
@ShengzhuPeng11 күн бұрын
Hi! I’m interested in a business collaboration. Could you please share your email Thanks
@Designing-hc5pz12 күн бұрын
when will you gonna release new model and it's about sd3 or sdxl?
@matheusreisfernandes846313 күн бұрын
When will the When will the pony diffusion be available?
@deefster7416 күн бұрын
why 1.5 vs XL? seems odd to release a new workflow on an old model
@bgtubber16 күн бұрын
Audio is too low. I could barely hear anything. :(
@ThinkDiffusion16 күн бұрын
So sorry about that! It'll be fixed next time we do a workflow video like this:) Hopefully you're still able to hear enough to try it out!
@bgtubber16 күн бұрын
@@ThinkDiffusion Yes, I had to crank it up and then it's audible. Thanks!
@Pellapoo18 күн бұрын
He named the workflow "Mein Kamph"
@MultiOmega191111 күн бұрын
bruh
@Cu-gp4fy20 күн бұрын
Thx for the workflow ! are there any other prepicessors u reco? noticed canny but not scribble
@ThinkDiffusion19 күн бұрын
Scribble is great too, especially with the xdog preprocessor!
@Mel-vc7csАй бұрын
Hello, I don't see all the sampling methods you are showing in this video. for example, I just see DPM++ 2M not DPM++ 2M Karras. Any idea why? Thanks
@Mel-vc7csАй бұрын
I don't have the roop option. can you explain why?
@rukakunАй бұрын
sorry if this is a stupid question, I'm a complete newbie. But, how does Think Diffusion differ from using stable diffusion? I'm (getting) familiar with Stable Diffusion, and am struggling to find what I can use ThinkDiffusion for other than generating images...
@user-ny2rv6kc7kАй бұрын
This is like the Paralympics
@MelihaMontajАй бұрын
0:15 isnt this paige from the game "5 dates"?
@MelihaMontajАй бұрын
@@dorian6021 even open AI's ceo cheats from real actors like scarlett, so...
@ShakkerAI_Ай бұрын
Paid Sponsorship & Business Consulting.May I get your e-mail?
@PunxTV1232 ай бұрын
is this your new channel sebastian?
@ThinkDiffusion2 ай бұрын
Hey! No this is a clip from ThinkDiffusion's course which is indeed recorded by Sebastian:) But this is not his channel specifically.
@aimademerich2 ай бұрын
Phenomenal
@nadaashry56022 ай бұрын
could you please do one video for room staging???
@ashishgupta10602 ай бұрын
Definitely awesome tutorial. we need more like this.
@ThinkDiffusion2 ай бұрын
Thank you Ashish, a lot more are coming every week!
@nithyaperiyasamy-cq9kw2 ай бұрын
To use comfyui, should we install the nodes and models in our machine??
@ThinkDiffusion2 ай бұрын
Hey! To use ComfyUI simply click it before starting a machine. We have 4 guides on how to use it here: learn.thinkdiffusion.com/tag/comfyui/
@nithyaperiyasamy-cq9kw2 ай бұрын
@@ThinkDiffusion thanks 👍
@Limitlazz3 ай бұрын
Maybe I'm blind but I can't find part 2?
@ThinkDiffusion3 ай бұрын
Don't worry you're not blind! We'll be uploading it later this week:)
@Limitlazz3 ай бұрын
@@ThinkDiffusion Happy to hear, thank you.
@jmcasler15123 ай бұрын
Thing Diffusion is the best out there! Amazing stuff!
@ThinkDiffusion3 ай бұрын
Thank you! We appreciate that 🫶
@chove934 ай бұрын
hi,how upload other model for controlnet (like ip-adapter-faceid-plusv2) ?
@chove934 ай бұрын
hi,how upload other model for controlnet (like ip-adapter-faceid-plusv2) ?
@chove934 ай бұрын
hi how instal model for control net ? like ip-adapter-faceid-plusv2
@DigiGrooveAI5 ай бұрын
I don't have ''disneyPixar....." in the options. What can I use for animation style like Pixar?
@ThinkDiffusion4 ай бұрын
You can download any model or LoRA from CivitAI or your computer. That means you can use any animation syle model from here www.civitai.com Just follow the step-by-step guide here: learn.thinkdiffusion.com/thinkdiffusion-faqs/#how-to-upload-a-model-by-url-eg-civitai
@chove935 ай бұрын
hi,how upload other model for controlnet (like ip-adapter-faceid-plusv2) ?
@muhammadbustamam46716 ай бұрын
can we upload our own LORAs?
@ThinkDiffusion4 ай бұрын
Yes, you can! Folow the below guide & just upload a file, instead of pasting a URL link:) Step-by-step process here: learn.thinkdiffusion.com/thinkdiffusion-faqs/#how-to-use-loras
@A_G4207 ай бұрын
what about loading comfyui workflows stored locally?
@jjda477 ай бұрын
What's the character limit for the text to image prompt? Is it really just 75 characters, or have you added more?
@ThinkDiffusion7 ай бұрын
Hey there! You have 75 tokens and the weighting is applied more to the first token, 1, 2, 3... etc. All the way to 75. On the 76th token it goes to a new batch and it will apply the same weighting as the 1st token. So the 76th token will indeed have a higher weighting for that specific batch. You can actually add more than 75 tokens or characters though. Hopefully this was helpful. :)
@jjda477 ай бұрын
@@ThinkDiffusion what happens if I exceed the number of characters or tokens per model?
@jsanchez7827 ай бұрын
Kinda of a newbie here, but 1000 for ns*w seems low? Why is it that much lower than the other ones?
@ThinkDiffusion7 ай бұрын
Hey there! No worries:) - we wanted to make a realistic model that also has nsfw capabilities, not a specifically nsfw model. A lot of the photorealistic custom models didn't include *any* nsfw pictures in their dataset.
@KIDNOFACE4 ай бұрын
down bad ai user
@christiandarkin7 ай бұрын
using it now - great model - well done
@ThinkDiffusion7 ай бұрын
Thank you Christian, that means a lot! We hope you'll have a great time using it:) Have a great day!
@christiandarkin7 ай бұрын
@@ThinkDiffusion my current fun workflow has your model linking straight into stable-video-diffusion to create alien wildlife films...
@ThinkDiffusion7 ай бұрын
haha that's awesome, you need to share those in our Discord!
@ashishgupta10608 ай бұрын
Best model in the market so far. Loved it <3
@ThinkDiffusion8 ай бұрын
Appreciate the kind words!
@Comic_Book_Creator8 ай бұрын
AnimateDiff, Deforum + ControlNet, TemporalKit, InfiniteZoom, SadTalker, Wav2Lip Studio and more ,, this are include on the RAPID A111111¡?
@ThinkDiffusion8 ай бұрын
Yes, these are all preinstalled on all 3 A1111 machines we offer - including the RAPID A1111 one:)
@Comic_Book_Creator8 ай бұрын
this is great, so what about the work and space.. also canyou add more payment options?
@ThinkDiffusion8 ай бұрын
Hi there! Yes, this was added to our backlog & we will soon support more payment options! Currently we are focusing on improving how the platform performs and other key things that are crucial for our community. Check back soon:)
@ThinkDiffusion8 ай бұрын
You can download ThinkDiffusionXL here: civitai.com/models/169868/thinkdiffusionxl Or you can try it out on our website on fast GPUs along with tons of other extensions & models preinstalled: www.thinkdiffusion.com Here are the prompt examples that you can try out: - Mid Shot Photo - Mid Shot Portrait - Architecture Photography - Wildlife Photography - Car Photography - Portrait (Female/Male) - Food Photography - Full Body Photo - Interior Photography - Landscape Photo/Photography - Hyperdetailed Photography/photograph - Close up portrait - Photograph/photography/photo - Cinematic - Epic Detailed - Distance Shot - Shot on (type camera here, for example Kodak, Hasselband, Sony etc.) For eyes just prompt the color of the eyes to get better looking eyes. If you don't do that you will probably end up with the SDXL Base Eyes which look more like marbles. So try: - Green Eyes - Blue Eyes - Brown Eyes - etc. You also can do more fancy colors like Aquamarin or Ruby. Thanks for watching & have a great day! :)
@Cu-gp4fy8 ай бұрын
thx looking good will give it a try!
@thinkdezign9 ай бұрын
How many models we can upload? is there any size limit? because models size is usally large around 2gb to 5gb or more
@ThinkDiffusion9 ай бұрын
Hey there! You can upload as many models as you want - there is no size limit. With the free plan you do get 10 GB of storage, but you already have 50+ popular models pre-installed... if you do want to download any other this should be enough to download 3-4 more. If that's not enough, you can always delete and reinstall any model in a matter of seconds (yes download speeds are that good) or you can subscribe to TD-PRO which gives you 200 GB of storage!
@thinkdezign9 ай бұрын
If we upload custom models from URL will they be available after we shutdown system and restart next day? i heard its a bUG in Think Diffusion - Is it Fixed Yet? TD-PRO comes with any free minutes or its just a add on package?
@ThinkDiffusion9 ай бұрын
Absolutely should be and are available! TD-PRO did come with free minutes as a promotion earlier but by default it's a 20% discount across all rates thanks for asking.
@AlexdeBrantes19 ай бұрын
Thanks for the simplified breakdown of LoRa's and embeddings!
@ThinkDiffusion9 ай бұрын
Glad it was helpful! :)
@ItsDigitalDreamer9 ай бұрын
This is a really informative video 🙂
@ThinkDiffusion9 ай бұрын
Glad you liked it Digital Dreamer!
@Cu-gp4fy9 ай бұрын
Love it 🔥🔥🔥
@ThinkDiffusion9 ай бұрын
<3
@ThinkDiffusion9 ай бұрын
Join our Discord here: discord.gg/XymB4Kb6Nm
@Cu-gp4fy9 ай бұрын
0:45 one punch man look out!
@ThinkDiffusion9 ай бұрын
AI is good but OPM is next level. Humans might be safe for just a bit longer
@ThinkDiffusion9 ай бұрын
Prompts for ChatGPT: 1. ‘’I am working on an apocalyptic AI movie and I need you to generate 5 different unique theme and look ideas. They should be simple enough to include them in every prompt I will use to generate initial pictures.’’ 2. ‘’I choose the (x number) theme and look. Now create a story for a 2 minute trailer, with 15 different frames. For each frame I want you to generate 5 keywords of the theme and 5 keywords of the look. Note that the theme and the look must stay consistent, while the story develops.’’ ‘3. ’Now create a voiceover that matches the frames story, theme, look & make it dramatic.’’ Thank you for watching! Let us know down below what videos or guides you want to see next! & don't forget to subscribe to learn more! 🚀
Пікірлер
Good video, I feel like the audio track of your voice could have been a bit louder compared to the background music.
Hi! I’m interested in a business collaboration. Could you please share your email Thanks
when will you gonna release new model and it's about sd3 or sdxl?
When will the When will the pony diffusion be available?
why 1.5 vs XL? seems odd to release a new workflow on an old model
Audio is too low. I could barely hear anything. :(
So sorry about that! It'll be fixed next time we do a workflow video like this:) Hopefully you're still able to hear enough to try it out!
@@ThinkDiffusion Yes, I had to crank it up and then it's audible. Thanks!
He named the workflow "Mein Kamph"
bruh
Thx for the workflow ! are there any other prepicessors u reco? noticed canny but not scribble
Scribble is great too, especially with the xdog preprocessor!
Hello, I don't see all the sampling methods you are showing in this video. for example, I just see DPM++ 2M not DPM++ 2M Karras. Any idea why? Thanks
I don't have the roop option. can you explain why?
sorry if this is a stupid question, I'm a complete newbie. But, how does Think Diffusion differ from using stable diffusion? I'm (getting) familiar with Stable Diffusion, and am struggling to find what I can use ThinkDiffusion for other than generating images...
This is like the Paralympics
0:15 isnt this paige from the game "5 dates"?
@@dorian6021 even open AI's ceo cheats from real actors like scarlett, so...
Paid Sponsorship & Business Consulting.May I get your e-mail?
is this your new channel sebastian?
Hey! No this is a clip from ThinkDiffusion's course which is indeed recorded by Sebastian:) But this is not his channel specifically.
Phenomenal
could you please do one video for room staging???
Definitely awesome tutorial. we need more like this.
Thank you Ashish, a lot more are coming every week!
To use comfyui, should we install the nodes and models in our machine??
Hey! To use ComfyUI simply click it before starting a machine. We have 4 guides on how to use it here: learn.thinkdiffusion.com/tag/comfyui/
@@ThinkDiffusion thanks 👍
Maybe I'm blind but I can't find part 2?
Don't worry you're not blind! We'll be uploading it later this week:)
@@ThinkDiffusion Happy to hear, thank you.
Thing Diffusion is the best out there! Amazing stuff!
Thank you! We appreciate that 🫶
hi,how upload other model for controlnet (like ip-adapter-faceid-plusv2) ?
hi,how upload other model for controlnet (like ip-adapter-faceid-plusv2) ?
hi how instal model for control net ? like ip-adapter-faceid-plusv2
I don't have ''disneyPixar....." in the options. What can I use for animation style like Pixar?
You can download any model or LoRA from CivitAI or your computer. That means you can use any animation syle model from here www.civitai.com Just follow the step-by-step guide here: learn.thinkdiffusion.com/thinkdiffusion-faqs/#how-to-upload-a-model-by-url-eg-civitai
hi,how upload other model for controlnet (like ip-adapter-faceid-plusv2) ?
can we upload our own LORAs?
Yes, you can! Folow the below guide & just upload a file, instead of pasting a URL link:) Step-by-step process here: learn.thinkdiffusion.com/thinkdiffusion-faqs/#how-to-use-loras
what about loading comfyui workflows stored locally?
What's the character limit for the text to image prompt? Is it really just 75 characters, or have you added more?
Hey there! You have 75 tokens and the weighting is applied more to the first token, 1, 2, 3... etc. All the way to 75. On the 76th token it goes to a new batch and it will apply the same weighting as the 1st token. So the 76th token will indeed have a higher weighting for that specific batch. You can actually add more than 75 tokens or characters though. Hopefully this was helpful. :)
@@ThinkDiffusion what happens if I exceed the number of characters or tokens per model?
Kinda of a newbie here, but 1000 for ns*w seems low? Why is it that much lower than the other ones?
Hey there! No worries:) - we wanted to make a realistic model that also has nsfw capabilities, not a specifically nsfw model. A lot of the photorealistic custom models didn't include *any* nsfw pictures in their dataset.
down bad ai user
using it now - great model - well done
Thank you Christian, that means a lot! We hope you'll have a great time using it:) Have a great day!
@@ThinkDiffusion my current fun workflow has your model linking straight into stable-video-diffusion to create alien wildlife films...
haha that's awesome, you need to share those in our Discord!
Best model in the market so far. Loved it <3
Appreciate the kind words!
AnimateDiff, Deforum + ControlNet, TemporalKit, InfiniteZoom, SadTalker, Wav2Lip Studio and more ,, this are include on the RAPID A111111¡?
Yes, these are all preinstalled on all 3 A1111 machines we offer - including the RAPID A1111 one:)
this is great, so what about the work and space.. also canyou add more payment options?
Hi there! Yes, this was added to our backlog & we will soon support more payment options! Currently we are focusing on improving how the platform performs and other key things that are crucial for our community. Check back soon:)
You can download ThinkDiffusionXL here: civitai.com/models/169868/thinkdiffusionxl Or you can try it out on our website on fast GPUs along with tons of other extensions & models preinstalled: www.thinkdiffusion.com Here are the prompt examples that you can try out: - Mid Shot Photo - Mid Shot Portrait - Architecture Photography - Wildlife Photography - Car Photography - Portrait (Female/Male) - Food Photography - Full Body Photo - Interior Photography - Landscape Photo/Photography - Hyperdetailed Photography/photograph - Close up portrait - Photograph/photography/photo - Cinematic - Epic Detailed - Distance Shot - Shot on (type camera here, for example Kodak, Hasselband, Sony etc.) For eyes just prompt the color of the eyes to get better looking eyes. If you don't do that you will probably end up with the SDXL Base Eyes which look more like marbles. So try: - Green Eyes - Blue Eyes - Brown Eyes - etc. You also can do more fancy colors like Aquamarin or Ruby. Thanks for watching & have a great day! :)
thx looking good will give it a try!
How many models we can upload? is there any size limit? because models size is usally large around 2gb to 5gb or more
Hey there! You can upload as many models as you want - there is no size limit. With the free plan you do get 10 GB of storage, but you already have 50+ popular models pre-installed... if you do want to download any other this should be enough to download 3-4 more. If that's not enough, you can always delete and reinstall any model in a matter of seconds (yes download speeds are that good) or you can subscribe to TD-PRO which gives you 200 GB of storage!
If we upload custom models from URL will they be available after we shutdown system and restart next day? i heard its a bUG in Think Diffusion - Is it Fixed Yet? TD-PRO comes with any free minutes or its just a add on package?
Absolutely should be and are available! TD-PRO did come with free minutes as a promotion earlier but by default it's a 20% discount across all rates thanks for asking.
Thanks for the simplified breakdown of LoRa's and embeddings!
Glad it was helpful! :)
This is a really informative video 🙂
Glad you liked it Digital Dreamer!
Love it 🔥🔥🔥
<3
Join our Discord here: discord.gg/XymB4Kb6Nm
0:45 one punch man look out!
AI is good but OPM is next level. Humans might be safe for just a bit longer
Prompts for ChatGPT: 1. ‘’I am working on an apocalyptic AI movie and I need you to generate 5 different unique theme and look ideas. They should be simple enough to include them in every prompt I will use to generate initial pictures.’’ 2. ‘’I choose the (x number) theme and look. Now create a story for a 2 minute trailer, with 15 different frames. For each frame I want you to generate 5 keywords of the theme and 5 keywords of the look. Note that the theme and the look must stay consistent, while the story develops.’’ ‘3. ’Now create a voiceover that matches the frames story, theme, look & make it dramatic.’’ Thank you for watching! Let us know down below what videos or guides you want to see next! & don't forget to subscribe to learn more! 🚀
Thank you for this! Really interesting 🙂
Appreciate it! 🔥