Machine Learning Engineering Open Book

AI Tensor Engine for ROCm

Data manipulation and transformation for audio signal processing, powered by PyTorch

GPU programming related news and material links

Fast and memory-efficient exact attention

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

Portfolio analytics for quants, written in Python

LeetGPU Solutions

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

⚡️ Lightning-fast backtesting engine to find your trading edge

Python Backtesting library for trading strategies

迅投QMT接口

[CVPR 2026] Official implementation of BiCo: Composing Concepts from Images and Videos via Concept-prompt Binding

A feature-rich command-line audio/video downloader

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3, Qwen3-MoE, DeepSeek-R1, GLM4.5, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Llava, …

Nsight Python is a Python kernel profiling interface based on NVIDIA Nsight Tools

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.5, DeepSeek-R1, GLM4.5, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Llava, Phi4, ...…

Train speculative decoding models effortlessly and port them smoothly to SGLang serving.

cuTile is a programming model for writing parallel kernels for NVIDIA GPUs

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

🤗 A PyTorch-native and Flexible Inference Engine with Hybrid Cache Acceleration and Parallelism for DiTs.

A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")

Complete solutions to the Programming Massively Parallel Processors Edition 4

LightTTS is a lightweight TTS inference framework optimized for CosyVoice2 and CosyVoice3, enabling fast and scalable speech synthesis in Python and supports stream and bistream modes.

MiMo: Unlocking the Reasoning Potential of Language Model – From Pretraining to Posttraining

slime is an LLM post-training framework for RL Scaling.

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

xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism