No video :(

NEW NSFW Pygmalion 7B IS HERE! THE Uncensored HOT Roleplay KING!

Подписаться 153 тыс.

Просмотров 143 тыс.

50% 1

Видео Поделиться Скачать Добавить в

The NEW Pygmalion 7B AI is an amazing open-source AI LLM model that is completely uncensored and fine-tuned for chatting and role-playing conversations! In this video, we dive into its features and compare it to the previous Pygmalion 6B model to find out which one is the best uncensored LLM roleplay model!
What do you think of Pygmalion 7B Ai? Let me know in the comments!
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
SOCIAL MEDIA LINKS!
✨ Support my work on Patreon: / aitrepreneur
⚔️ Join the Discord server: bit.ly/aitdiscord
🧠 My Second Channel THE MAKER LAIR: bit.ly/themake...
📧 Business Contact: theaitrepreneur@gmail.com
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
JOIN MY DISCORD SERVER : / discord
Watch my previous Oobabooga Video for the Webui installation: • UPDATED TextGen Ai Web...
Previous Pygmalion video: • NSFW Pygmalion AI! Unc...
Pygmalion 7B model: huggingface.co...
Metharme 7B model: huggingface.co...
Tavern Ai: github.com/Tav...
Metharme yaml file: mega.nz/file/j...
Booru: pastebin.com/7...
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
►► My PC & Favorite Gear:
i9-12900K: amzn.to/3L03tLG
RTX 3090 Gigabyte Vision OC : amzn.to/40ANaue
SAMSUNG 980 PRO SSD 2TB PCIe NVMe: amzn.to/3oBR0WO
Kingston FURY Beast 64GB 3200MHz DDR4 : amzn.to/3osdZ6z
iCUE 4000X - White: amzn.to/40y9BAk
ASRock Z690 DDR4 : amzn.to/3Amcxph
Corsair RM850 - White : amzn.to/3NbXlm2
Corsair iCUE SP120 : amzn.to/43WR9nW
Noctua NH-D15 chromax.Black : amzn.to/3H7qQSa
EDUP PCIe WiFi 6E Card Bluetooth : amzn.to/40t5Lsk
Recording Gear:
Rode PodMic : amzn.to/43ZvYlm
Rode AI-1 USB Audio Interface : amzn.to/3N6ybFk
Rode WS2 Microphone Pop Filter : amzn.to/3oIo9Qw
Elgato Wave Mic Arm : amzn.to/3LosH7D
Stagg XLR Cable - Black - 6M : amzn.to/3L5Fuue
FetHead Microphone Preamp : amzn.to/41TWQ4o
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Special thanks to Royal Emperor:
- Totoro
Thank you so much for your support on Patreon! You are truly a glory to behold! Your generosity is immense, and it means the world to me. Thank you for helping me keep the lights on and the content flowing. Thank you very much!
#GPT4 #GPT3 #ChatGPT #pygmalion
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
WATCH MY MOST POPULAR VIDEOS:
RECOMMENDED WATCHING - All LLM & ChatGPT Video:
►► • CHATGPT
RECOMMENDED WATCHING - My "Tutorial" Playlist:
►► bit.ly/TuTPlay...
Disclosure: Bear in mind that some of the links in this post are affiliate links and if you go through them to make a purchase I will earn a commission. Keep in mind that I link these companies and their products because of their quality and not because of the commission I receive from your purchases. The decision is yours, and whether or not you decide to buy something is completely up to you.

Опубликовано:

23 авг 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 423

@Aitrepreneur Год назад

HELLO HUMANS! Thank you for watching & do NOT forget to LIKE and SUBSCRIBE For More Ai Updates. Thx

@corvonerodellatempes Год назад

Is there a way to use it from mobile browser like with the older model?

@chopchop3458 Год назад

Love your stuff!! It has been very helpful for me! BUT HOW DO YOU HAVE VOICES?? I WANT VOICES TOO!

@Muckihito Год назад

Noob here. Can i run this on my android phone? And if yes, how?

@MrKill-wg8vk Год назад

Could you tell me if there are any good LLMs for coding I know people want certain things and you want to make videos about those certain things but it would be great if you could help

@Tys_Black Год назад

How did you get those Ai voices?

@AI_EmeraldApple Год назад

AI stuff is moving SO fast I can hardly keep up even with the great youtube content creators posting updates all the time! Thank you for making these great videos, keeping us up to date.

@qaesarx Год назад

This is just the TIP OF THE ICEBERG, the AI race is ON, it will end in a disaster OR we will merge with AI, both things are TERRIFIYING!!!!! AHHHHHH 😀

@AI_EmeraldApple Год назад

@@qaesarx I'm personally an AI maximalist. I kinda feel like at this point if you don't use it, you will be left behind.

@qaesarx Год назад

@@AI_EmeraldApple Behind ? Youre done. AI is the future, the question is... what happens in the end. There is always an end. 😀

@touma-san91 Год назад

In recent years it definetly has moved fast.. But I also remember how bad the first chatbots used to be, compared to AI, it's very hard to call them even bots anymore

@gaberoch9153 Год назад

It was a very nice touch of using the AI generated voice to add more "role play" to the example. Nice work!

@Avalon19511 Год назад

was tts enabled in Tavern?

@DihelsonMendonca Год назад

How he did that ? What program he used for tts ?

@touma-san91 Год назад

@MUSIC EDITOR TM No. It is not. But SillyTavern has TTS now if you clone the repository instead of download release.

@touma-san91 Год назад

@@Avalon19511 No. But SillyTavern has TTS now if you clone the repository instead of download release.

@Sloff1155 Год назад

Pygmalion-7B 404 not found.

@Hodoss Год назад

12:55 LOL how desperate Pygmalion tries to keep you from leaving: "A... A... A WILD MUGGER APPEARS!!!"

@yukicandra561 Год назад

With such AI, I think we'll eventually see a game with incredibly interactive NPCs that can even conduct NSFW conversations with those side characters. as long they're connected to the database server.

@Galaxia53 Год назад

I was just thinking that with AI powered NPCs it would hurt a lot more if the character got abducted and you had to save her. And then I thought what if the player connected with the character and then the NPC gets killed off. That just sounds cruel

@Galaxia53 Год назад

@@N1ghtR1der666 There are a lot of bots on youtube but I think that I can identify a lot of them

@seabass-nj2qb Год назад

Skyrim has it, they added chat gpt

@Robin-jk3uz Год назад

@@seabass-nj2qb they did and it's impressive

@QWERTY-gp8fd Год назад

@@Galaxia53 that shit is cool idea for mmorpg game.

@yahifumeno Год назад

This man uses imagery by AI, probably uses chatgpt to make his commentary better and who knows, he might even be AI himself!

@gamerdre7376 Год назад

@@clumsy_en cybernetic hivemind. We are the borg....

@Aitrepreneur Год назад

Shhh!🤫🤫🤫

@liminalai Год назад

Our overlord😍

@fixelheimer3726 Год назад

.. with a very good french accent simulator too ;D

@jasonhemphill6980 Год назад

@@fixelheimer3726doesn't say "uhhhhh" enough

@moor6451 Год назад

doing a Oobabooga+LangChain !!!!!!!. I know I'm not the only one who would be thrilled to see it! . Please consider making this video happen, we're all eagerly waiting for a video on chatting with PDF in webui sounds like a fantastic idea. peace from Morocco

@gwky Год назад

When they make a 13B or 30B model, then we'll be having a lot more fun. I don't really see that much of a difference going from the 6B to the 7B llama quantized version, maybe it can keep a slightly more context-relevant conversation, but it seems that we still need to wait for million token context research paper to be turned into viable code, so that we can truly have deep and meaningful text adventures or conversations without the AI forgetting what it was talking about a minute ago.

@Galaxia53 Год назад

It may take a while for such big models to be able to run locally

@gwky Год назад

@@Galaxia53 true, I'd recommend Google's VMs since it's probably the best pay for what you use PaaS. I've been using the Nvidia L4 VMs, would recommend, especially if your GPU at home can't run these models well.

@michaelbone6894 Год назад

The million token context research paper is about carrying over a small amount of information over 1M tokens. It can't use 1M tokens of context. For that, probably the only viable way would be to add on a separate memory module that can store and retrieve information using traditional computer science approaches. Computers have way better memory then humans anyway, it just needs to be integrated into a full language model, with the core LLM being just one component.

@zengrath Год назад

I agree, the 7b models no matter how amazing they are for 7b just can't compare to 13b and 30b models. their outputs are always super simplistic and short compared to larger models and makes using them boring. I am more interested in story telling then chat bot so maybe I just don't understand the appeal of this model.

@Djorgal Год назад

@@michaelbone6894 Yeah, that's basically about how to efficiently summarize 1M token worth of context into something that fits within the 4k token that can actually be used as context. There's going to be losses. But... that's fine. That's also basically how humans work too. We don't remember word for word every conversations we had so far with someone, what we have in our head is a running summary of how our previous interactions went. If the AI is capable to keep a running summary as context, that's fine.

@life2030 Год назад

Where do those realistic voices come from? Are those voice built-on or generated by text-to-speech please? If it is an external text-to-speech program, then how do you setup please? Thank you!

@KypHeM Год назад

Elevenlabs text to speech

@Ez-se2dl Год назад

What kind of AI did you use to voice the text? Or was it a real human?

@traveleractual 9 месяцев назад

The huggingface link to this model is returning a 404

@zygerman7703 Год назад

the Pygmalion 7B model is gone... I see 404 not found. Now what?

@IchEsseHaende Год назад

Could you make a video about how to set the parameters (Ai response configuration)? Both Pygmalion 6b and 7b perform really weird for me and I have no idea what these sliders actually do and what configuration is good for what purpose.

@user-xl2lw4sv9v Год назад

I would like this video as well

@rachelsales1164 Год назад

Up!

@Insight_Matters Год назад

yes, most dialoges seem to be rubbish and make no sense...if I increase temperture or extend minimum tokens it gets worse and worse. I wished there would be ever just one dialogue as coherent as the one in your vid :D

@IchEsseHaende Год назад

@@Insight_Matters same for me and I get endless repetitions really fast

@touma-san91 Год назад

I haven't had much of issues with 6B but 7B is definetly really broken.

@gustavdreadcam80 Год назад

Good video, I will try out the new pygmalion model immediatly. I think the old model did it's job great but coherency is one of it's main problems and the new 7b seems to do it better now. Did you use elevenlabs for the text to voice generation?

@WilliamFoxdev Год назад

Also super curious about the TTS (text to speech) that was used here. It was pretty good.

@Beetlebomb3D Год назад

Thank you for titling the video with uncensored! Helps a lot

@maverick9860 Год назад

The pygmalion 7b link isnt working

@crysishk Год назад

Recently, I have been pretty good at typing with one hand. Looks like I will get even better.

@TomCoolery 10 месяцев назад

the pygmalion 7b isn't available anymore? the link goes to a 404 page.

@mohammedaslam2912 Год назад

I like how it describes the mood and expression at the start of every reply. Can we somehow hook it up with an image generator that automatically animates the character's face based on this ?

@oldpernilong0 Год назад

I think sillytavern has a algorythm that simulate character expressions by simulating how their dialogue match with emotions and choose a image that you set that match the emotion. Like if you have a character with happy/neutral png files they will appear like a dating sim when the character is happy or neutral.

@user-xl2lw4sv9v Год назад

yo that would be epic

@CannedPsycho- Год назад

The developer of Sillytavern developed extras for the interface that allows you to place images labeled something like joy.png inside a character folder and have it appear every time the AI thinks that's what the character is feeling.

@touma-san91 Год назад

@@CannedPsycho- Yup, he did. But you need to have 6 png's of the character with different expressions so it's not really what Mohammed asked as what he wants would use way more VRAM.

@sophytes1430 Год назад

Does anyone know what happened to this model? apparently he was removed from hugging face :(

@traveleractual 9 месяцев назад

This pygmalion 7b model link returns a 404 error.

@miguelgargallo Год назад

you are awesome and helped a lot of people

@Nepoxification Год назад

Nice vid as always! How did you achieve the tts? Did you do that in editing or is that some extenion?

@momentomori1747 Год назад

The new 7b model sounds a lot more coherent. I'll have to test it out tonight as a replacement for shygmalion.

@Legnog822 Год назад

tonight

@momentomori1747 Год назад

@@Legnog822 It was a pain to get working with KoboldAI, but it is really good. Definitely will be using it going forward.

@cr1tikal_arc Год назад

@@momentomori1747 how did you get it to be good at all? i tried using and it's extremely repetitive and can never maintain an understanding of the context of the current conversation for like more than 1 message. it's really fast compared to other models i've used but that just isn't worth it when it can't give me a reply actually related to the conversation. sometimes it'll even reintroduce itself randomly.

@momentomori1747 Год назад

@@cr1tikal_arc Haven't had issues with maintaining context or reintroductions, but I do see what you mean by repeating itself. I typically go in and edit the bot's replies every so often to avoid it repeating itself. (ie If the character is prone to smiling edit out all references in previous messages so the bot doesn't start smiling five times in a single reply.) I've been meaning to test out increasing the repetition penalty as well to see if that helps. I think it is worse than shygmalion was with regards to being prone to repeating itself.

@denischons9246 Год назад

@@momentomori1747 did you have the problem of not identifying the model type? How did you fixed it?

@dersnorri7187 Год назад

The Pygmalion Link seems to be dead now. I tried to locate it manually and found the model, but I now have an error because the model has no config file.

@mygamecomputer1691 Год назад

Once again thank you my AI overlord! You are the hero we don’t deserve.

@rmirabelle Год назад

I cannot even manage to run Ooba, which fails to start. Errors all over the place, all completely cryptic. Tried one-click installers from multiple sources - all fail for different reasons. This is dependency hell, where every library silently depends on a thousand other libraries which are out of date before you even start.

@user-ml3qv8go5d Год назад

KoboldAI is meant for text adventure type scenarios so you missed not running Metharme with it. It would perhaps perform better.

@Aitrepreneur Год назад

Ah good point! That probably why!

@tdfilmstudio Год назад

Which voiceover tool have you used for the video dialogues? Can you please share

@bahriaproperties1143 Год назад

I second this how did he managed to get text to speech

@SpudHead42 Год назад

Thirded.

@Whisper_InThe_Rain Год назад

what did you use for the text to speech was it bark? I love the emphasis on the words.

@adamait5219 Год назад

May I Ask what you used to voice over the chats? That sounds nice!

@Galaxia53 Год назад

He said it's Elevenlabs

@MrArrmageddon Год назад

I want it too. But I suspect it's not free.

@adamait5219 Год назад

@@Galaxia53 oh, sorry, I missed it mb. The good elevenlabs voices are paywalled as far as I remember, right?

@Galaxia53 Год назад

@@adamait5219 He said it in the comments somewhere I meant. I figured if I just said Elevenlabs I wouldn't sound as credible. I don't know anything about it my self. Was planning on looking into it but if they're paywalled then that could be pretty bad.

@Galaxia53 Год назад

Pygmalion 13b has been released

@ThisOrThat13 Год назад

We're slowly making our way to AI chat, voice, and video all mixed into one! We'll never need to leave the house again. Audio from Bark dropped into Adobe? That's a lot of editing.

@deepsmoo8955 Год назад

How do you get text-to-speech narration and character speaking?

@mungrutsai Год назад

is elevenlabs?

@roseydeep4896 Год назад

Two questions: How can I create my own character for this webui and is the voiceover some kind of built in feature? Also I kind of thought the 6b model was a little inconsistent, weird and leaning *toooooooooooooooooooooooo much* into nsfw stuff....... so yeah. I'm definitely switching. Anyway thanks for the vid :) Have a nice one!

@musa3344 Год назад

I must say your casting is amazing 😂😂

@VongolaChouko Год назад

Page to specific model link deleted?

@noobbonexd Год назад

hey! i am completly new into ALL of this. i used Chat GPT a few times before and thats... it... it would be great iof you could make a introduction videos for people like me to slowly get us into what even happens here before maybe going into more detail, it would be appreicated a lot for sure

@German4Media Год назад

Thanks for the video! Are you actually changing something at TavernAI directly? For example the temperature or something?

@lemlibloodaxe313 Год назад

Dont know if its a problem with this model or not but it starts out great for the first 20 mins then it seems to just continue to repeat itself over and over and I cant do much to change its replies. I've had it repeat *giggles* like 100 times in its reply and nothing changes unless you start a new chat with your character then its fine again for a while. I tried it with at least 2 different char cards and they end up eventually doing the same.

@touma-san91 Год назад

It's more of a problem with the model and token limits. 6B suffers lot less from that issue

@sch117sch Год назад

Great! Thanks for the update!

@jinjurbreadman Год назад

the link to pygmalion 7b in the video description doesn't work...

@ThmSphone Год назад

Regarding TavernUI: I use SillyTavern instead of TavernUI. It's prettier, and has a few options that can help increase text longevity / AI memory, like creating a history and backlog where you can save key info using a very small amount of tokens. However, after a couple months using both Taverns, I found that the lack of streaming (showing every letter being generated on a text) is a huge letdown, and now I go 50/50 on using SillyTavern and the regular oobabooga UI.

@bgill7475 Год назад

Yep, SillyTavern is better.

@Trahloc Год назад

The folks making TavernAI should just import all of SillyTavern. Having two dev teams is unfortunate.

@IchEsseHaende Год назад

Is it normal that the chat in sillytavern is way slower than when I'm using oobabooga directly, or am I doing something wrong? In oobabooga the generation starts almost instantly and in sillytavern I wait between 20-40 seconds.

@Trahloc Год назад

@@IchEsseHaende --no-stream which is recommended, massively slows down chat compared to direct oob. From what I've noticed it requires the full 2000 tokens to be calculated even if the LLM responds with just "Sure!". It's one of the reasons I use direct most often lately but I still really like Silly.

@IchEsseHaende Год назад

@@Trahloc that's what it feels like for me to, even short answers need a really long time. Do you think this will be fixed?

@chyldstudios Год назад

Why are there only 24 hours in a day? I need at least 100 hours/day to play with all these toys!

@Drakmour666 Год назад

Was those audio lines also made by AI from previous videos or it was recorded by humans? It sounds too good...

@bsmithril Год назад

I've noticed after so many messages it starts to forget where it is, repeat itself a lot, and completely change in personality.

@julx97 Год назад

Are you planning on making new videos on every new LLM model that is coming out?

@Aitrepreneur Год назад

Only the most interesting ones

@StillNight77 Год назад

@@Aitrepreneur I'll watch a comparison video for every new model if the changes are noticeable. Just sayin'.

@RomiWadaKatsu Год назад

The whole conversation between K and a bratty freeloader teenager was an absolute fever dream that I didn't know I wanted but I enjoyed way more than I thought

@VIRGILIANSHAILER Год назад

the link to Pygmalion 7B is no longer working. 404 error

@reinanoliveira3127 Год назад

Is anyone experiencing this error? Found the following quantized model: models\gozfarb_pygmalion-7b-4bit-128g-cuda\pyg7b-4bit-128g.safetensors ?

@3drocket Год назад

How did you do the character voice overs?

@gianluca3131 Год назад

When downloading the model I get the error: Traceback (most recent call last): File “D:\ChatAI\TextGenWebUI\text-generation-webui\server.py”, line 189, in download_model_wrapper links, sha256, is_lora = downloader.get_download_links_from_huggingface(model, branch, text_only=False) File “D:\ChatAI\TextGenWebUI\text-generation-webui\download-model.py”, line 103, in get_download_links_from_huggingface fname = dict[i][‘path’] KeyError: 0 Any idea?

@cryptogaming9935 Год назад

your oobabooga version is outdated, i had the same error. After re-downloading and re-installing oobabooga it went perfectly. Of course doing that might screw up usage of older language models you might have installed.

@xuonivVi Год назад

idk if i'm stupid or what i'm doing wrong, but "The model returned empty message" keeps popping up everytime i try to chat. i've been at it for hours but i can't find a solution

@ezbaisalgado4169 Год назад

I been trying to download the Pygmalion 7B mode and it just keeps on giving me the same error code, I tried to manually downloaded but it still does not work, can somebody tell me what em I doing wrong? the error code is ''fname = dict[i][‘path’] KeyError: 0''

@cryptogaming9935 Год назад

i have the same problem, no idea edit: i solved it, i had to install my web ui oobabooga from scratch, now it works

@kaiio5639 Год назад

I don't think it matters at all if the model is uncensored or not. As you showed in your Vicuna video, all models adapt to the prompt, so if the character card is spicy, then the conversation with it will also be spicy. Also while EVAI refused to generate such content in Oobabooga's webui, there was no problem in SillyTavern, I didn't even need to modify the character information, only converted it to json. Even if it didn't work, I could have just modified the Character Bias and set a low insertion depth (making it take more priority) to make any character card spicy. Write something like "[System note: {{char}} is eager to engage in spicy roleplay.]"

@touma-san91 Год назад

You don't even have to do all of that. You just need to slowly push the direction of discussion towards NSFW.

@d.r8892 Год назад

I wish you would make a single video explaining how to install all the instances to make the AI work. It is a bit hard jumping videos to understand what you are saying.

@lifson22 Год назад

yeah its honestly a huge mess at this point. Not coherent

@d.r8892 Год назад

@@lifson22 ah I am glad I am not the only one

@Djorgal Год назад

What set of parameters give the best outcomes for the model?

@thephaerohassasin2869 Год назад

of all the recent models you posted which one do you think is the best

@veryseriousperson_ Год назад

Whats the AI Voice that was used?

@prixmalcollects9332 Год назад

same question

@sgramstrup Год назад

If it is good at role-playing then i would imagine it would be great for langchain/autogpt and such (keeps in character). Can you check that next time ?

@TheRealBlackNet Год назад

Strange the 6B model is "funnier" as the 7B model or I need other characters. In direct one by one same character the answers in 7B are shorter and start to repeat them self often if you hit continue. 7B also waits for the user to start something while 6B contiunes more freely. On the other hand 6B drifted often in a horror action bunker with gov plot, that did I not get with 7B till now.

@biggestdog420 Год назад

Thanks!

@marcthenarc868 Год назад

Those voices 😶 ... where did you get them ?

@Daydricoldsmith Год назад

I'm assuming it was elevenlabs.

@0verride881 Год назад

Is there any version that ummm.. can send you images that you set beforehand? From my understanding, there are only chatbots that cannot send you anything beyond text. Surely making an image repository of some sort for the bot to send contextually would be possible?

@dibbidydoo4318 Год назад

How the hell do you have 'Download custom models or LoRA' in your webui, I don't have that.

@dedhatab4423 Год назад

There is no pygmalion models on your link. Also, other 7B pygmalion models dont work.

@dorsal1157 Год назад

There is no google collab? :(

@DihelsonMendonca Год назад

How did you do Text to speech TTS on this example ??? 😮

@CYkillone Год назад

How much vram do I need to run the 7b model?

@electrocola9765 Год назад

Coolest thing is that this GUI use all my GPUs (2) and I can select amount of VRAM for the model. What a time

@Drakmour666 Год назад

How long can it continue the dialogue? Does it have a system that summarizes all the conversation to know what was before it exceeded in tokens?

@user-yy6wz9wc1b Год назад

ok this is getting out of control and it's incredibly spectacular

@davidnajafi3949 Год назад

What is the text to speech used?

@famberlight Год назад

How did you make the voiceover? Was it done during editing of video or in taverai?

@prixmalcollects9332 Год назад

same question

@damienk7311 Год назад

Where is that character browser interface in Tavern coming from for you? Tavern looks nothing like that for me when I download it.

@thebestofthequest7486 Год назад

is there a google colab for this/can you alter the old one to use the new model?

@gio3061 Год назад

There is cutting edge, bleeding edge, and now we have a bleeding border.

@knarfamoranemix6030 Год назад

A problem with instructions is that they rarely ever work. Then you have to look for other instructions elsewhere, but they don’t work either and if different sets of instructions are incmpatible, they are hard to combine.

@amirk7955 Год назад

thanks a lot! can you please make a google doc for this and the Tavern Ai just like you did before? my computer is super weak but with the google doc it worked well, thank you.

@SpikyBlade Год назад

You mean a google colab ?

@Aitrepreneur Год назад

you can try using this one: colab.research.google.com/drive/18L3akiVE8Y6KKjd8TdPlvadTsQAqXh73#scrollTo=S8MSUddUpA1s

@enesahmetoglu8151 Год назад

Thank you so much man so muh

@IMajst3RI Год назад

@@Aitrepreneur It only does local links for me so I can't get it. Any help?

@amirk7955 Год назад

@@Aitrepreneur thank you for you effort you are the best! but i still cannot make it work - i managed to get to the ogaboga things but unable to make API to paste in the tavern ..... can you make one with the KoboldAI thing? thank you

@chompychomp13 Год назад

The link to the model brings up a 404 error

@dezmont9191989 Год назад

I'm waiting for time when AI can create Visual (video even better) for text games. Hopefully one day, AI will be able to make complex games in Unreal Engine.

@BrennonBaxter Год назад

this is actually built in function within the koboldai "new ui", do some searching but it offers up a stable diffusion instance that can be triggered when you click generate, it will take the setting/description of the the current chat and generate an image.

@gamerdre7376 Год назад

Reminds me of text based games from the early 80s.

@necrobard Год назад

Had the 6B one working fine from your previous guide but I'm having way too many errors on this one. Couldn't download the model through the webui, so I download it manually, then can't select the model. Started over from scratch using your previous guide was able to download the model this time, but still couldn't select it without forcing into the llama type, then once oobabooga finally loads the model, it's completely retarded in TavernAI, just repeating itself over and over again. Feels like some step or other configuration is missing.

@sircatsbythe3rd Год назад

I'm having the same issue and don't know what's wrong

@denischons9246 Год назад

How did you managed to force the model type?

@necrobard Год назад

@@denischons9246 3:41 He covers it there, if you did a new / fresh install of oobabooga you need to install another model first to actually launch the web ui to be able to do that.

@YoureNowOnTV Год назад

I keep thinking that these are visual art models ..... my brain is not programming itself correctly 😌

@metasamsara Год назад

No google doc version this time? I can't get oobabooga thing to run on my laptop. I'm thinking it's a driver issue with APU architecture.

@lucisac6405 Год назад

Is this the best 7b model suitable for 8gb vram that is uncensored?

@Kazic5000 Год назад

@Aitrepreneur Hey if someone develops one of those google docs to run the ai like in the last one, and you find out about it, could you please link it in the description?

@JakeBananas Год назад

how can i use this model with python calls?

@adamwalid876 Год назад

Great video! is there a way to make this work on a phone? i can set it up on my computer but i dont know how to make it work on mobile as well

@danielalexander9418 Год назад

What are you using to create the voices?

@iPasqualo Год назад

Great video, thanks! What is the best story teller model? (Uncensored)

@mahdiyarabbasi3800 Год назад

Help I tried to install this model according to this video and the video (UPDATED TextGen Ai WebUI Install! Run LLM Models in MINUTES!) but I encountered this problem (ModuleNotFoundError: No module named 'llama_inference_offload'). Likewise, after downloading this model (GPT4 x Alpaca), I encountered this problem. I am using CPU mode

@kaiio5639 Год назад

5:30 Have you tried SillyTavern? TavernAI only looks pretty without providing much functionality. I prefer Oobabooga's interface over it because it lets me have more control over the conversation, but SillyTavern finally convinced me to make the switch. I mostly use the character bias functionality, but it can do much more, with its most prominent feature, I believe, being group conversations.

@suhrayuh Год назад

Could you please explain the settings to me? People are always saying that SillyTavern is better but a lot of the settings are kind of intimidating 😭

@zaidankreshnandi255 Год назад

I cant open the pygmalion link what happend? Anf it said when in tavern ai there no connection

@ion_propulsion7779 Год назад

How did you do the text to speech?

@TomiTom1234 8 месяцев назад

Can you please tell me, where is that "api" that I have to check to be able to connect it with TavernAI ? After the new interface of "oobabooga" I can't find it under "interface mode" section.. PLEASE HELP!!

@SimonsCatSE 8 месяцев назад

same

@GrieferGamer Год назад

Is the voiceover made with elevenlabs?

@tylermamerow Год назад

I did everything just like in the video, but the characters do not respond to me after I post something.

@TheDogn Год назад

I'm getting an error that I think means my GPU isn't beefy enough. I tried running it on CPU memory but I'm getting errors about pytorch and cuda, which ChatGPT tells me is required for GPU inference. Is it possible to run Pygmalion with CPU?

Далее

NEW 13B WizardLM UNCENSORED LLM! A NEW Beginning for AI MODELS?!

15:12

NEW 13B WizardLM UNCENSORED LLM! A NEW Beginning for AI MODELS?!

Просмотров 67 тыс.

5 Things Sociopaths Do

12:03

5 Things Sociopaths Do

Просмотров 8 млн

НИКИТА СДЕЛАЛ ПРЕДЛОЖЕНИЕ ЛЕРЕ? 😍

01:00

НИКИТА СДЕЛАЛ ПРЕДЛОЖЕНИЕ ЛЕРЕ? 😍

Просмотров 193 тыс.

Девочки, у вас тоже так? 💅🏻✨

00:17

Девочки, у вас тоже так? 💅🏻✨

Просмотров 29 тыс.

You can Run but you can't Hide 🤣😍 #funny #dubai #snatcher #viral #fyp #dubai #india #tajikistan

00:28

You can Run but you can't Hide 🤣😍 #funny #dubai #snatcher #viral #fyp #dubai #india #tajikistan

Просмотров 43 тыс.

Коля Киргиз жив?! Новое видео из Санкт-Петербурга шокирует криминальный мир!

01:00

Коля Киргиз жив?! Новое видео из Санкт-Петербурга шокирует криминальный мир!

Просмотров 191 тыс.

Etiquette Expert Answers Etiquette Questions From Twitter | Tech Support | WIRED

11:31

Etiquette Expert Answers Etiquette Questions From Twitter | Tech Support | WIRED

Просмотров 1,7 млн

Next level AI art Control | My workflow

23:02

Next level AI art Control | My workflow

Просмотров 97 тыс.

NEW Flux and NF4 Flux for ComfyUI. FASTEST FLUX! BEST AI MODEL! How to use Flux.

17:02

NEW Flux and NF4 Flux for ComfyUI. FASTEST FLUX! BEST AI MODEL! How to use Flux.

Просмотров 558

This Llama 3 is powerful and uncensored, let’s run it

14:58

This Llama 3 is powerful and uncensored, let’s run it

Просмотров 129 тыс.

CREATE NSFW Roleplay Ai Characters! Pygmalion & Tavern AI SECRETS!

22:41

CREATE NSFW Roleplay Ai Characters! Pygmalion & Tavern AI SECRETS!

Просмотров 206 тыс.

Everything wrong with Woke Feminism!

17:10

Everything wrong with Woke Feminism!

Просмотров 7 млн

RIP Stable Diffusion! BEST FREE UNCENSORED AI Model Is HERE!

16:26

RIP Stable Diffusion! BEST FREE UNCENSORED AI Model Is HERE!

Просмотров 147 тыс.

The difference between afraid and annoying

11:47

The difference between afraid and annoying

Просмотров 723 тыс.

INSTALL BEST UNCENSORED Roleplay TextGen UI LOCALLY in 1 CLICK!

27:49

INSTALL BEST UNCENSORED Roleplay TextGen UI LOCALLY in 1 CLICK!

Просмотров 164 тыс.

Google Is a Monopoly: Judge Rules Against Tech Giant in Landmark Antitrust Case

10:47

Google Is a Monopoly: Judge Rules Against Tech Giant in Landmark Antitrust Case

Просмотров 67 тыс.

НИКИТА СДЕЛАЛ ПРЕДЛОЖЕНИЕ ЛЕРЕ? 😍

01:00

НИКИТА СДЕЛАЛ ПРЕДЛОЖЕНИЕ ЛЕРЕ? 😍

Просмотров 193 тыс.