Explorations into the proposed SDFT, Self-Distillation Enables Continual Learning, from Shenfeld et al. of MIT

A template for research projects in computer science/machine learning using python and julia

Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support

Training library for Megatron-based models with bidirectional Hugging Face conversion capability

For releasing code related to compression methods for transformers, accompanying our publications

[ICML 2024] Official Implementation of SLEB: Streamlining LLMs through Redundancy Verification and Elimination of Transformer Blocks

Compressed LLMs for Efficient Text Generation [ICLR'24 Workshop]

A CLI to estimate inference memory requirements for Hugging Face models, written in Python.

Agent skills for Manim to create 3Blue1Brown style animations.

Experimental mini python jupyter kernel

A Model Agnostic function to directly remove specified layers from the LLM

Official implementation of the ICLR paper "Streamlining Redundant Layers to Compress Large Language Models"

Official repository for EMNLP2025 paper "IG-Pruning: Input-Guided Block Pruning for Large Language Models"

Setup guide for ML training on NVIDIA DGX Spark (GB10 Blackwell, CUDA 13, aarch64)

Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

Extending the Context of Pretrained LLMs by Dropping Their Positional Embedding

LLM checkpointing for DeepSpeed/Megatron

nanoRLHF: from-scratch journey into how LLMs and RLHF really work.

Use git from python, fast

Solar vs GLM vs Phi

🦋 An Infographic Generation and Rendering Framework, bring words to life with AI!

A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

Data mapping framework for rust stuff

Tooling for exact and MinHash deduplication of large-scale text datasets