Your channel is new to me and was recommended by algorithm. I don't see any reviews of chinese models or really open source ones like Intellect1 first non-corporate one. I see most videos here are corporate commercial marketing stuff.
you can change the audio track to germany if you want, and that's i think the reason why UA-cam autotranslate the title to germany (your country i supposed)
You can run it easily locally with 128Gb RAM (70B in Q8 usually takes 90Gb), but original Llama is always too censored to the level of "database", there can't be any glimpse of intellect. Chinese already on the same level but censored even weirdly-Qwen 32B decline to instruct Llama installation.
Great insight on the new model for sure. My GTX 1080 6GB VRAM should be okay with it. What was the name of the app/site you were using to make a mockup react app?
Knowledge cut off Dec 2023 is what makes me disappointed. Thought of download but cancel. Meta is filthy rich it can generously give us latest information like Oct 2024 but dont 😢
Thanks for mentioning the artificialanalysis, such a comprehensive analysis for choosing the best API and model.
They do an excellent job
New sub, no bs. Just news. Ty!
Straight to the point, I like it. What is the environment you are using to get artifacts on an open source model?
+1
+1
+1
Your channel is new to me and was recommended by algorithm. I don't see any reviews of chinese models or really open source ones like Intellect1 first non-corporate one. I see most videos here are corporate commercial marketing stuff.
What do you use for screen recording? The quality is amazing!
Is there any place to check how the model behaves while consuming resources from the hardware?
this is hard to find!
how to set the one language in llama3.3-70B?
I FIND THIS TOOL YOU ARE WORKING ON INTERESTING, DO YOU MAKE IT AVAILABLE FOR DOWNLOAD?
Hey can you make a video where we can use OLLAMA on a free GPU on cloud ?
why the hell is the complete video description and the title in german and the speaker is talking english?
Google just began trying to auto translate videos - I am not sure if I have control over this. Do others see this?
you can change the audio track to germany if you want, and that's i think the reason why UA-cam autotranslate the title to germany (your country i supposed)
You can run it easily locally with 128Gb RAM (70B in Q8 usually takes 90Gb), but original Llama is always too censored to the level of "database", there can't be any glimpse of intellect. Chinese already on the same level but censored even weirdly-Qwen 32B decline to instruct Llama installation.
Llama can be 100% uncensored, you just need the right system prompt.
can you tell me how is your mouse movement so clean and smooth?
Probably just a low DPI (sensitivity) so he moves his hand a lot to get a little distance
@@Dark_Souls_3 looks like screen studio
@@GamesGVGits a mac feature from what i know
Can run locally on MacBook M4Max 64 GB
Not sure but try with something like LM studio or else
I dont know if I Feel comfortable having anything from Meta installed all over my pc... lol.
Fireworksai & Groq need to add it
Great insight on the new model for sure. My GTX 1080 6GB VRAM should be okay with it. What was the name of the app/site you were using to make a mockup react app?
6gb too small for llm
Knowledge cut off Dec 2023 is what makes me disappointed. Thought of download but cancel. Meta is filthy rich it can generously give us latest information like Oct 2024 but dont 😢
please drink some water
Your mom needs to drink some water
@@elcoollow5955 ok. now go to school kevin.
When is elon releasing his llm model?
First
Huh, common Indian trait.
I know this is unrelated but give the Quran a read, also have a good day :)
I did 1 / 10. The guy that has within the Quran is a psychopath!
And the bible too. We can always use more atheists :)
@@XHackManiacXcompletely unrelated, but the Dune too..
@@XHackManiacX lmao, great reply