Breakthrough: Run Massive Models On Any Device (ex: LLaMA 65b)

ByEditor

Update: Sorry for the audio sync issue ?

In this video, we talk about Petals. A new project combines old-ish technology with large language models to allow you to run even the largest models in a distributed fashion on any device. This incredible new implementation truly decentralizes LLMs (LLaMA, Bloom, MPT, etc) and allows consumer-grade computers to run any large model.

Enjoy 🙂

Become a Patron ? – https://patreon.com/MatthewBerman
Join the Discord ? – https://discord.gg/xxysSXBxFW
Follow me on Twitter ? – https://twitter.com/matthewberman
Subscribe to my Substack ?? – https://matthewberman.substack.com

Links:
Petals – https://petals.dev/
Research – https://research.yandex.com/blog/petals-decentralized-inference-and-finetuning-of-large-language-models
Petals Google Colab – https://colab.research.google.com/drive/1uCphNY7gfAUkdDrTx21dZZwCOUDCMPw8?usp=sharing

Similar Posts

  • GEMINI 1.0 Beats GPT-4 | AlphaCode 2 is better than 90% of coders | Gemini Nano on Pixel Phones!

    More Videos Diving Deep Into All the Announcements:(coming Soon) LINKS MENTIONED:https://twitter.com/demishassabis/status/1732416480741118179https://storage.googleapis.com/deepmind-media/gemini/gemini_1_report.pdfhttps://www.wired.com/story/google-deepmind-demis-hassabis-gemini-ai/https://blog.google/technology/ai/google-gemini-ai/https://cloud.google.com/tpu?hl=enhttps://deepmind.google/discover/blog/competitive-programming-with-alphacode/https://storage.googleapis.com/deepmind-media/AlphaCode2/AlphaCode2_Tech_Report.pdfhttps://deepmind.google/technologies/gemini/https://bard.google.com/chat Get on my daily AI…

  • AI Mega Update: New Tools and Huge News

    New ai tools for creativity, productivity, and business. ChatGPT plugins, Google AI, humanoid robots, and so much more. AI news keeps speeding up.

    *the audio cuts off in the last 10-15 seconds because the song got flagged for copyright. Google and Taito partnered to release an ar version of space invaders that’s coming out in July.

    ——————————————————————————————–

    ? AI video effects with Kaiber: https://kaiber.ai/?via=obscurious
    ? Use AI to build your website with 10web: https://10web.io/?_from=obscurious
    ? Level up on YouTube with VidIQ: https://vidiq.com/obscurious

    ? Twitter: https://twitter.com/ObscuriousMind

    ——————————————————————————————–

    AI Tools and News mentioned:
    Jizai Arms: https://jizai-arms.com/
    Sanctuary AI: https://sanctuary.ai/
    Tesla Optimus: https://twitter.com/Tesla_Optimus/status/1658576897490530305
    Amazon Burnham: https://twitter.com/rowancheung/status/1659651124821893123
    Google Music: https://aitestkitchen.withgoogle.com/
    DragGAN: https://huggingface.co/papers/2305.10973
    Invideo: https://invideo.io/?ref=obscuriousmind
    *use the code OM25 to receive a 25% discount on invideo
    OpenAI Shap e: https://github.com/openai/shap-e
    Nyric: https://lovelacestudio.com/
    Unreal Engine 5.2: https://twitter.com/UnrealEngine/status/1656678238159269895?
    10web: https://10web.io/?_from=obscurious
    Coca Cola Ad: https://youtube.com/watch?v=VGa1imApfdg&
    ChatGPT NPC: https://youtube.com/watch?v=U4W2rGH9oWs&

    ——————————————————————————————–

    My camera setup:
    ? Camera: https://amzn.to/3xm9EnU
    ? Lights: https://amzn.to/3v7Dcmx
    ? Tripod: https://amzn.to/3KsYbXm
    ? Microphone: https://amzn.to/3jrZ6eK

    ——————————————————————————————–

    Chapters:
    0:00 Intro
    0:19 Jizai Arms
    0:40 Sanctuary AI
    1:09 Tesla
    1:34 Amazon
    2:03 Google music
    2:31 DragGAN
    3:02 invideo
    4:29 Midjourney 5.1
    4:43 Open AI Shape-e
    4:51 Nyric
    5:11 Unreal Engine 5.2
    5:26 10web
    6:22 Coca Cola Stable Diffusion
    6:59 Wendy’s AI
    7:28 ChatGPT NPC simulation
    7:40 ChatGPT plugins
    9:28 ChatGPT web browsing
    9:43 Google
    10:55 Space Invaders AR