-
Level AI
- India
- https://shreyansh26.github.io
- @shreyansh_26
Highlights
Lists (4)
Sort Name ascending (A-Z)
Starred repositories
Standalone commandline CLI tool for compiling Triton kernels
Learn CUDA Programming, published by Packt
Minimalistic 4D-parallelism distributed training framework for education purpose
flash attention tutorial written in python, triton, cuda, cutlass
a minimal cache manager for PagedAttention, on top of llama3.
Building blocks for foundation models.
📰 Must-read papers and blogs on Speculative Decoding ⚡️
🦀 Small exercises to get you used to reading and writing Rust code!
Efficient Deep Learning Systems course materials (HSE, YSDA)
My learning notes/codes for ML SYS.
[NeurIPS 2024] Fast Best-of-N Decoding via Speculative Rejection
Recipes to train reward model for RLHF.
Official PyTorch implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden States
For optimization algorithm research and development.
📄 A curated list of awesome .cursorrules files
An AutoHotKey script for Windows that lets a user change virtual desktops by pressing CapsLock + <num>.
Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.
Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.
Universal LLM Deployment Engine with ML Compilation
Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.
A Bulletproof Way to Generate Structured JSON from Language Models
A bibliography and survey of the papers surrounding o1
AI powered one-click comprehensive docs from transcripts and text.