From bare metal to a 70B model: infrastructure set-up and scripts https://
bit.ly/45XsNfi
#AI #MachineLearning #DeepLearning #LLMs #DataScience
HARDWARE
-
Infrastructure Setup and Scripts for 70B Model Deployment
By
–
-
QLoRA Bug Consumes 4888GB VRAM in Llama 3 405B
By
–
One bug in bitsandbytes caused QLoRA to consume an extra *read notes* 4888 GB of VRAM. Even with QLoRA, fine-tuning Llama 3 405B on long sequence lengths ain't cheap.
-
Energy-Efficient AI Inference Acceleration Platform Recognized by Lazard
By
–
Honored to be named to @Lazard
's VGB AI Infra 40! Our energy-centric #AI #inference #acceleration addresses the critical need for efficient AI solutions. Proud to stand with fellow pioneers, driving sustainable, high-performance AI #computing forward. https://
lazard.com/research-insig
hts/venture-growth-banking-insights-lazard-vgb-ai-infra-40/
… -
Running 7B Parameter Models Locally on Apple Silicon
By
–
I am more and more excited about Local ML on Apple Silicon. Core ML in particular is starting to be a super nice stack to build on. Question: Do you want to have a 7B parameter model running at 30+ tokens/second, using less than 4GB of memory on your Mac? Then you need to
-
Programmable Block from Space Engineers Enables Custom Code Execution
By
–
Amazed by my new real-world Programmable Block from Space Engineers, built by my colleague @AIRobin_net !
— Marek Rosa | European🇪🇺 | South African🇿🇦 (@marek_rosa) 22 juillet 2024
I can play Pong, view pictures, and even upload new code.@SpaceEngineersG pic.twitter.com/sii6n5IOTYAmazed by my new real-world Programmable Block from Space Engineers, built by my colleague @AIRobin_net ! I can play Pong, view pictures, and even upload new code. @SpaceEngineersG
-
Open Source Transforming AI and Hardware Innovation
By
–
Watch @jimkxa
, @RajaXg , Chris Walker of @UntetherAI and Charlie Cheng, Board Advisor for @Andes_Tech talk about how #OpenSource is transforming AI and hardware: -
Nvidia Develops B20 AI Chip for Chinese Market Compliance
By
–
@nvidia is developing a new AI chip, "B20," for the Chinese market! This move aims to comply with US export regulations while maintaining a solid presence in China’s AI sector. Will this help Nvidia stay ahead of local competitors like @Huawei
? #AI #Nvidia #Innovation #Tech -
OpenAI’s AI Chip and Latest AI Developments Today
By
–
Top stories in AI today: -OpenAI wants its own AI chip
-The state of AI humanoids and robotics
-How to test and customize GPT-4o mini
-Apple’s new 7B open-source AI model
-6 new AI tools & 4 new AI jobs Read more: http://
therundown.ai/p/openai-plans
-its-own-ai-chip
… -
Large-scale 3D printing revolutionizes rapid house construction
By
–
Todos conocemos las impresoras 3D capaces de reproducir pequeños objetos pero, qué pasaría si la impresora fuera los suficientemente grande?
— Juan Merodio (@juanmerodio) 22 juillet 2024
Pues que te imprimen una casa en tiempo récord 😳 pic.twitter.com/SIoaPDnGMRTodos conocemos las impresoras 3D capaces de reproducir pequeños objetos pero, qué pasaría si la impresora fuera los suficientemente grande? Pues que te imprimen una casa en tiempo récord
-
Soccer Team Roles Analogy for Tech Roles and Responsibilities
By
–
Similar analogies but in soccer:
– Goalie: Hardware folks
– Defender: Infra & data engineers
– Midfielder: All-around SWE and ML engineers
– Offense: General AI researchers
– Striker: Star AI researcher training the biggest models and doing yolo runs
– Coach: people managers