Open-source implementation of AlphaEvolve

A Datacenter Scale Distributed Inference Serving Framework

dlBLAS: clean and efficient kernels

NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmer…

Build, manage, and scale your AI agents with ease.

MiroThinker is open-source agentic models trained for deep research and complex tool use scenarios.

Mamba SSM architecture that supports training on variable-length sequences

MiroRL is an MCP-first reinforcement learning framework for deep research agent.

MiroTrain is an efficient and algorithm-first framework for post-training large agentic models.

Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation

InternEvo is an open-sourced lightweight training framework aims to support model pre-training without the need for extensive dependencies.

A library to analyze PyTorch traces.

A Quirky Assortment of CuTe Kernels

FlagGems is an operator library for large language models implemented in the Triton Language.

Efficient Triton Kernels for LLM Training

KernelBench: Can LLMs Write GPU Kernels? - Benchmark with Torch -> CUDA problems

Tensors and Dynamic neural networks in Python with strong GPU acceleration

The Python programming language

CUDA Templates for Linear Algebra Subroutines

FlashInfer: Kernel Library for LLM Serving

veRL: Volcano Engine Reinforcement Learning for LLM

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Throughput-oriented multi-turn inference engine for KernelBench [ICML '25]

A debugging and profiling tool that can trace and visualize python code execution

A PyTorch native platform for training generative AI models

PyTorch native quantization and sparsity for training and inference