Hugging Face
Models
Datasets
Spaces
Community
Docs
Enterprise
Pricing
Log In
Sign Up
3
4
2
SagarDesai
sagarsdesai
Follow
21world's profile picture
JamesBentley's profile picture
hassoudi's profile picture
4 followers
Β·
4 following
SDcodehub
AI & ML interests
AGI, AI, RL, LLM, ML, MLOPS
Recent Activity
upvoted
an
article
about 2 months ago
Introducing Training Cluster as a Service - a new collaboration with NVIDIA
reacted
to
merve
's
post
with π
about 2 months ago
stop writing CUDA kernels yourself we have launched Kernel Hub: easy optimized kernels for all models on Hugging Face π₯ use them right away! it's where the community populates optimized kernels π€ this release comes in three parts > Kernel Hub: contains (as of now) 14 kernels > kernels: Python library to load kernels from Kernel Hub > kernel-builder: Nix package to build kernels for PyTorch (made using PyTorch C++ frontend) when building models, your regular workflow should be pulling kernels from Hub and building your model with them π€ here's a practical example with RMSNorm: 1. pull the kernel from Hub with `get_kernel` 2. decorate with `use_kernel_forward_from_hub` 3. inject it to your model we'd love to hear your feedback! ππ» we also welcome kernel contributions by community π₯Ήπ - request kernels here: https://huggingface.co/spaces/kernels-community/README/discussions/1 - check out this org: https://huggingface.co/kernels-community - read the blog: https://huggingface.co/blog/hello-hf-kernels
upvoted
an
article
3 months ago
Visualize and understand GPU memory in PyTorch
View all activity
Organizations
sagarsdesai
's Spaces
1
Sort:Β Recently updated
Runtime error
make_over
π