Alexey Dertev's picture

2 8 41

Alexey Dertev

d8rt8v

·

AI & ML interests

None yet

Recent Activity

liked a model about 13 hours ago

OuteAI/Llama-OuteTTS-1.0-1B-GGUF

liked a model 3 days ago

bartowski/deepcogito_cogito-v1-preview-qwen-32B-GGUF

reacted to onekq's post with 😎 3 days ago

We desperately need GPU for model inference. CPU can't replace GPU. I will start with the basics. GPU is designed to serve predictable workloads with many parallel units (pixels, tensors, tokens). So a GPU allocates as much transistor budget as possible to build thousands of compute units (Cuda cores in NVidia or execution units in Apple Silicon), each capable of running a thread. But CPU is designed to handle all kinds of workloads. CPU cores are much larger (hence a lot fewer) with branch prediction and other complex things. In addition, more and more transistors are allocated to build larger cache (~50% now) to house the unpredictable, devouring the compute budget. Generalists can't beat specialists.

View all activity

Organizations

None yet

d8rt8v's activity

upvoted a paper 7 days ago

Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems

Paper • 2504.01990 • Published 13 days ago • 232

upvoted an article 7 months ago

Article

Llama can now see and run on your device - welcome Llama 3.2

Sep 25, 2024

• 188

upvoted an article 8 months ago

Article

SmolLM - blazingly fast and remarkably powerful

Jul 16, 2024

• 352

upvoted 2 collections 9 months ago

abliterated-v3

Latest gen of the abliterated models I've produced • 17 items • Updated Jun 3, 2024 • 117

Saiga

LLaMA-based or Mistral-based Russian chat models • 9 items • Updated Mar 4 • 29