GitHub directory, page 11

Utilities intended for use with Llama models.

An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

Open Lakehouse Format for Multimodal AI. Convert from Parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, and PyTorch with more integrations coming..

CoreNet: A library for training deep neural networks

LiteRT-LM is Google's production-ready, high-performance, open-source inference framework for deploying Large Language Models on edge devices.

Efficient Triton Kernels for LLM Training

Chronos: Pretrained Models for Time Series Forecasting

SWE-bench: Can Language Models Resolve Real-world Github Issues?

Lemonade helps users discover and run local AI apps by serving optimized LLMs right from their own GPUs and NPUs. Join our discord: https://discord.gg/5xXzkMu8Zk

Gemma open-weight LLM library, from Google DeepMind

Core ML tools contain supporting tools for Core ML model conversion, editing, and validation.

On-device AI across mobile, embedded and edge for PyTorch

The RedPajama-Data repository contains code for preparing large datasets for training large language models.

Achieve state of the art inference performance with modern accelerators on Kubernetes

The Abstraction and Reasoning Corpus

PyTorch code and models for VJEPA2 self-supervised learning from video.

A unified inference and post-training framework for accelerated video generation.

A machine learning compiler for GPUs, CPUs, and ML accelerators

Set of tools to assess and improve LLM security.

Genome modeling and design across all domains of life

Official repository for the Boltz biomolecular interaction models

Open-source framework for the research and development of foundation models.

Everything about the SmolLM and SmolVLM family of models

A lightweight, powerful framework for multi-agent workflows and voice agents

AllenAI's post-training codebase

An advanced 3D Gaussian Splatting renderer for THREE.js

The best OSS video generation models, created by Genmo

[MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

LiteRT, successor to TensorFlow Lite. is Google's On-device framework for high-performance ML & GenAI deployment on edge platforms, via efficient conversion, runtime, and optimization

Waymo Open Dataset

Code for the paper "Evaluating Large Language Models Trained on Code"

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

LLMRouter: An Open-Source Library for LLM Routing

The hub for EleutherAI's work on interpretability and learning dynamics

Holistic Evaluation of Language Models (HELM) is an open source Python framework created by the Center for Research on Foundation Models (CRFM) at Stanford for holistic, reproducible and transparent evaluation of foundation models, including large language models (LLMs) and multimodal models.

A simple, performant, and scalable Jax LLM!

Fully open data curation for reasoning models

τ-Bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

Developer Asset Hub for NVIDIA Nemotron — A one-stop resource for training recipes, usage cookbooks, datasets, and full end-to-end reference examples to build with Nemotron models

Agentic RL Training at Scale

Home of StarCoder2!

Chai-1, SOTA model for biomolecular structure prediction

Reference implementations of MLPerf® training benchmarks

PyTorch building blocks for the OLMo ecosystem

Humanity's Last Exam

Reference implementations of MLPerf® inference benchmarks

Data and tools for generating and inspecting OLMo pre-training data.

A benchmark built to evaluate and improve agent capabilities for supporting legal work.

DataComp for Language Models

The HIP Environment and ROCm Kit - A lightweight open source build system for HIP and ROCm

Pytorch-based framework for solving parametric constrained optimization problems, physics-informed system identification, and parametric model predictive control.

An alignment auditing agent capable of quickly exploring alignment hypothesis

A clean, modular SDK for building AI agents with OpenHands V1.

Qualcomm® AI Hub Models is our collection of state-of-the-art machine learning models optimized for performance (latency, memory etc.) and ready to deploy on Qualcomm® devices.

Generative AI with ONNX Runtime

A JAX-based simulator for autonomous driving research.

Build plugins that feel like native, first-class features of ChatGPT.

GPU accelerated decision optimization

Low-bit LLM inference on CPU/NPU with lookup table

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

Measuring and evolving with the frontier of agent work

Train Models Contrastively in Pytorch

Code for the Molmo2 Vision-Language Model

Backward compatible ML compute opset inspired by HLO/MHLO

Our inference and training framework to run on the Cosmos Models

Find issues worth your attention.