verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

Train transformer language models with reinforcement learning.

An agentic-first and HuggingFace-native RL framework for research (9k lines).

DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms

Scalable toolkit for efficient model reinforcement

An LLM post-training framework with vLLM for RL Scaling

Agentic RL on Any Harness at Scale

A safetensors extension to efficiently store sparse quantized tensors on disk

DFlash: Block Diffusion for Flash Speculative Decoding

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM5.3, Gemma4, Llava, P…

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

Ongoing research training transformer models at scale

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…

Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).

3x Faster Inference; Unofficial implementation of EAGLE Speculative Decoding

Reproduction of DeepSeek-R1

Contrib repository for the OpenTelemetry Collector

An open-source framework for detecting, redacting, masking, and anonymizing sensitive data (PII) across text, images, and structured data. Supports NLP, pattern matching, and customizable pipelines.

Context OpenTelemetry Collector processor

The Triton backend for the ONNX Runtime.

Splits Keras with Tensorflow backends into two or more submodels.

DSPy: The framework for programming—not prompting—language models

LLM training in simple, raw C/CUDA

ONNXMLTools enables conversion of models to ONNX

onnxruntime-extensions: A specialized pre- and post- processing library for ONNX Runtime

Triton Model Analyzer is a CLI tool to help with better understanding of the compute and memory requirements of the Triton Inference Server models.

🕹️ Performance Comparison of MLOps Engines, Frameworks, and Languages on Mainstream AI Models.

Set of tools to assess and improve LLM security.

Master programming by recreating your favorite technologies from scratch.

Microsoft Automatic Mixed Precision Library