Stars
LUPINE is a GPU over IP bridge allowing GPUs on remote machines to be attached to CPU-only machines.
Doing simple retrieval from LLM models at various context lengths to measure accuracy
Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.
A library with extensible implementations of DPO, KTO, PPO, ORPO, and other human-aware loss functions (HALOs).
The official Porsche Design System repository, offering fundamental UXI guidelines and a library of reusable web components to enable designers and developers to build consistent, intuitive, and hi…
A Heterogeneous Benchmark for Information Retrieval. Easy to use, evaluate your models across 15+ diverse IR datasets.
The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.
[EMNLP 2022] Training Language Models with Memory Augmentation https://arxiv.org/abs/2205.12674
Supercharge Your LLM Application Evaluations 🚀
A dataset of 100,000 minutes of driving video compressed using a VQ-VAE.
A Data Streaming Library for Efficient Neural Network Training
A high-throughput and memory-efficient inference and serving engine for LLMs
The Memory Layer for AI Agents - Drop-in memory infrastructure for AI agents and apps. Context that persists. Built for production.
Write scalable load tests in plain Python 🚗💨
Implementation of Flash Attention in Jax
The official implementation of “Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training”
A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.
Tevatron - Unified Document Retrieval Toolkit across Scale, Language, and Modality. Demo in SIGIR 2023, SIGIR 2025.