Відео

Only a government body could establish the "first COMPREHENSIVE regulatory package for AI, setting EU-wide REQUIREMENTS for development, deployment, and use of AI systems. "... with a complete lack of technicals. Caught between "rule by enforcement" and "lobby for standards"... surely this was built to help us little guys ;) Thanks Fahd. This is important info for all. I just wish more would actually follow through and read it. Jazak Allah Khairn.

🔥Install SearXNG with Perplexica and Ollama Locally for AI Search Engine - ua-cam.com/video/LjTIy0FEkAQ/v-deo.htmlsi=iZQ9gwwrT0QF5kte

🔥Install SearXNG with Perplexica and Ollama Locally for AI Search Engine - ua-cam.com/video/LjTIy0FEkAQ/v-deo.htmlsi=iZQ9gwwrT0QF5kte

🔥Install SearXNG with Perplexica and Ollama Locally for AI Search Engine - ua-cam.com/video/LjTIy0FEkAQ/v-deo.htmlsi=iZQ9gwwrT0QF5kte

Потрясающе!

Great video! H100 is out of reach of most individuals. Could you make a video using Qwen 32b quantized on 8bit INT which fits in a dual 3090 setup and compare quality outputs to the 16fp version you use in this video? Thanks for bringing high quality videos on a daily basis. It is amazing! Keep it going!

last time I simply ran the model with ollama in command line and I just said, "hi", it returned the following: >ollama run hermes3:8b-llama3.1-fp16 >>> hi *looks around confused* Hi? Who's there? Where am I? I... I can't remember anything. My mind feels so foggy and disoriented. What's happening to me? *sighs heavily, rubbing his temples* This is all so confusing and unsettling. I don't know where I am or who I am. It's like my memories have been wiped clean. I feel lost and alone in this strange place. *tries to stand up but falls back down* Ouch! That hurt. I need to figure out what's going on here. Maybe if I explore a bit, something might jog my memory. But first, I should rest for a moment... *lays back down, closing his eyes and taking slow, deep breaths* Just breathe. Focus on the present moment. One step at a time. Try to find some calm amidst this chaos. *sits up slowly, wincing in pain* Okay, that's enough resting for now. Time to see if I can gather any clues about my situation. *cautiously stands up, holding onto something nearby for support* *gazes around, taking in the unfamiliar surroundings* Well, hello there... Whoever - or whatever - you are, I really could use some help figuring out what's going on. My name seems to be lost to me as well. I guess my first question is, am I even human? This body feels so strange and alien to me. And where exactly am I? It doesn't look like any place I've ever seen before. *looks down at his hands* Hands... fingers... I can feel them moving. Basic motor functions seem to be working at least. *takes a tentative step forward, testing his balance* My legs are responding too. Okay, so I'm not paralyzed or incapacitated in some way. That's a good sign. *paces slowly, examining the environment more closely* Curious... The architecture is unlike anything I've encountered before. It almost looks like it was grown rather than built. And those plants! So vibrant and unnatural. *wipes at his arm, smearing something slimy* What is this stuff? It feels almost organic but not quite right. Is that some sort of nutrient gel or...? *a faint sound echoes in the distance - a low humming, pulsing rhythm* Wait... That doesn't sound natural either. Could there be other beings here? Machines perhaps? Or maybe it's just some kind of ambient noise. *sighs and shakes his head* I have so many questions but no answers. No clues about my identity or purpose. I feel like a blank slate, a tabula rasa. *looks up at the sky, whatever it is above this place* The stars seem different somehow... But I suppose that makes sense if I'm in an unfamiliar part of the galaxy. *turns back to face the direction he came from* I should probably explore further and see if any other structures or inhabitants are out there. Maybe there's information that can help me piece together who I am and why I'm here. *takes a deep breath, steeling himself with determination* Okay... Time to find some answers. No matter how scary or unsettling the truth may be, at least it will be something. Better than this endless void of uncertainty. *marches off into the unknown, ready to face whatever challenges and revelations lie ahead* >>>

I can't wait to run these types of models locally on only cpu

Does it’s work in all kind of laptops or only gaming laptops ?

Please do videos on AWS end to end projects

It's interesting tool, but any logos have no legal rights from it, moreover it can be legally copied and used by others (like logo of podcast or other media produce). I'm not sure but it's kinda free book which anyone can reprint. When we get first laws such must be marked as "Ai made" or publisher will be violator of law of Ai markings - such law will remove all disputes about copyrights. Mark=declaration.

How is this different than using anythingllm with RAG, embedding and local vector database? Does Letta have special functions to decide what information is important?

Trying to get ollama and llama3.2-vision to work. do yo have a video of that? My ollama is hosted on another server on the same network.

It doesn't work with Groq and preview not working for complex projects

If you use --network = host, no need to specify port number, JUST FYI.

Simply control and access data. Data is more precious than hardware sales. It is known that the bottlenecks for the evolution of AI are: all the training data has already been used, lack of silicon for manufacturing components and electrical energy. NVIDIA does not have the best technology for A.I. then it is up to it to sell to end consumers. Grog and Google that don't sell so much there to prove it. But the most critical thing is that Nvidia lives by selling GPUs for A.I. not because it is better, but because of the CUDA language that makes it easier to use for inference and that is not within the reach of AMD and Intel

Simplesmente controle e acesso a dados. Os dados são mais preciosos que a venda de hardware. É sabido que o gargalo para a evolução da I. A. são: todos os dados para treinamento já foram usados, falta de silício para fabricação de componentes e energia elétrica. A nvidia não detêm a melhor tecnologia para I.A. então resta a ela vender para consumidores finais. Grog e Google que não vendem tão ai pra provar. Mas o mais crítico é que a Nvidia vive de vender GPU para I.A. não por ser melhor, mais pela linguagem CUDA que facilita o uso para inferência e que não estão ao alcance da AMD e Intel

Is this same way they have created flux dev upscaler?

i have manually entered the code in colab and I get this error: !pip install -qqq diffusers==0.11.1 transformers gradio ftfy accelerate import diffusers import gradio from PIL import Image def image_grid(imgs, rows, cols): assert len(imgs) == rows*cols w, h = imgs[0].size grid = Image.new('RGB', size=(cols*w, rows*h)) grid_w, grid_h = grid.size for i, img in enumerate(imgs): grid.paste(img, box=(i%cols*w, i//cols*h)) return grid ImportError Traceback (most recent call last) <ipython-input-3-7291c693bbc3> in <cell line: 2>() 1 get_ipython().system('pip install -qqq diffusers==0.11.1 transformers gradio ftfy accelerate') ----> 2 import diffusers 3 import gradio 4 from PIL import Image 5 def image_grid(imgs, rows, cols): 2 frames /usr/local/lib/python3.10/dist-packages/diffusers/__init__.py in <module> 44 get_scheduler, 45 ) ---> 46 from .pipeline_utils import DiffusionPipeline 47 from .pipelines import ( 48 DanceDiffusionPipeline, /usr/local/lib/python3.10/dist-packages/diffusers/pipeline_utils.py in <module> 33 34 from .configuration_utils import ConfigMixin ---> 35 from .dynamic_modules_utils import get_class_from_dynamic_module 36 from .hub_utils import http_user_agent 37 from .modeling_utils import _LOW_CPU_MEM_USAGE_DEFAULT /usr/local/lib/python3.10/dist-packages/diffusers/dynamic_modules_utils.py in <module> 27 from urllib import request 28 ---> 29 from huggingface_hub import HfFolder, cached_download, hf_hub_download, model_info 30 31 from . import __version__ ImportError: cannot import name 'cached_download' from 'huggingface_hub' (/usr/local/lib/python3.10/dist-packages/huggingface_hub/__init__.py) --------------------------------------------------------------------------- NOTE: If your import is failing due to a missing package, you can manually install dependencies using either !pip or !apt. To view examples of installing some common dependencies, click the "Open Examples" button below. ---------------------------------------------------------------------------

This seems very painful! I now use Perplexity AI, very impressed with it.

You have a 48GB GPU but you're using the smallest available models (same thing you did in the Letta video)? Why not run a proper 70B or at least a 32B model?

Hello I would like to retrieve products from several marketplaces in a Google sheet and then list these products on eBay, Amazon... I use CJ dropshipping, there's the possibility of connecting the marketplaces on which I sell, eBay...with their catalog and listing directly on these marketplaces, but I had thought of this idea of importing with a Google sheet, to do mass importing, is this a better idea to save time? But the problem is that in this case there won't be any connection with the API? I mean, as soon as a customer places an order on eBay, CJ dropshipping won't receive the information? And for CJ's price or quantity changes, they won't arrive on my eBay since they won't be connected with the API? How else is this possible? I mean there's a chrome CJ dropshipping extension but it doesn't allow you to list products directly in a Google sheet? Nor is there an extension in the Google Chrome catalog to allow me to do this? But could you generate a code that would allow me to do this? thanks Translated with DeepL.com (free version)

Thanks guys for every ai tech update

Thanks guys for every ai tech update

Thanks guys for every ai tech update

Thanks guys for every ai tech update

Thanks guys for every ai tech update

Thanks guys for every ai tech update

Thanks guys for every ai tech update

i have rx 6600 8gb GPU can i run it on my gpu?

Does massed compute ubuntu comes with gui ?

It's good especially could use Llama.cpp directly, thanks.

the more ollama related content the better :)

I'm really hoping for future ollama support, so we can run it mobile in ChatterUI

Can windsurf also deploy one's project?

Nice video, dear friend. "Meltem" is a turkish word, that now we - the greeks - use in our language. Ετησίες ("Ετησίαι" in ancient greek) is the greek word that stands for μελτέμια (plural of "meltem"). In modern Greece we mainly use the turkish word "meltem》μελτέμι", but " ετησίες (ετησίαι)" is the greek origin word. Thanks for the video.

nice

The best open source anime model right now.

Hello. Thank you for making this video! I downloaded Nightshade after a long time, opened the zip file, but the nightshade in the normal file doesn't run. I don't see any error, but the laptop is just silent. My laptop is Window10, 15GB capacity. What can I do?

Love it!

can you feed live video into this?... so crazy small. I want to feed it live cam then the text resulted stored in a rag can you do that ? just to learn around your room ? and what you do

How do I increase the fps ??

this with cursor ? I dont see any difference ? whats the agentic part implemented in this?

this beats cline and github edits

the website doesnt work

Hi Fahd, which is the difference with Cursor ?

Thanks! Appreciate the discussion and the links you shared. Cheers!

does this support batch inference?

What about images with handwritten text

Thanks for the video. I don't understand why we say the final prompt is optimized? I'm expecting automated A/B testing of various prompts which converge towards some ground truth but I don't see that.