Raphael-Hao

Raphael-Hao's Stars

• deepseek-ai/DeepSeek-V3

  Language:Python99.3k74161216.2k

• astral-sh/uv

  An extremely fast Python package and project manager, written in Rust.

  Language:Rust67.6k1267.6k2k

• geekan/MetaGPT

  🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming

  Language:Python54.4k8298246.5k

• deepseek-ai/awesome-deepseek-integration

  Integrate the DeepSeek API into popular softwares

  33.8k3821283.8k

• stanfordnlp/dspy

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

  Language:Python28.2k1801.3k2.2k

• waydabber/BetterDisplay

  Unlock your displays on your Mac! Flexible HiDPI scaling, XDR/HDR extra brightness, virtual screens, DDC control, extra dimming, PIP/streaming, EDID override and lots more!

  27k792.7k477

• kvcache-ai/ktransformers

  A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations

  Language:Python15k1117791.1k

• deepseek-ai/FlashMLA

  FlashMLA: Efficient MLA kernels

  Language:C++11.7k10264899

• modelscope/ms-swift

  Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 500+ LLMs (Qwen3, Qwen3-MoE, Llama4, GLM4.5, InternLM3, DeepSeek-R1, ...) and 200+ MLLMs (Qwen2.5-VL, Qwen2.5-Omni, InternVL3.5, Ovis2.5, Llava, GLM4v, Phi4, ...) (AAAI 2025).

  Language:Python9.9k443.3k870

• deepseek-ai/3FS

  A high-performance distributed file system designed to address the challenges of AI training and inference workloads.

  Language:C++9.3k86207938

• deepseek-ai/DeepEP

  DeepEP: an efficient expert-parallel communication library

  Language:Cuda8.5k81275925

• deepseek-ai/open-infra-index

  Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation

  7.9k4790282

• meta-llama/llama-stack

  Composable building blocks to build Llama Apps

  Language:Python7.7k128434979

• deepseek-ai/DeepGEMM

  DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling

  Language:Cuda5.7k55110694

• NVIDIA/DALI

  A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

  Language:C++5.5k921.7k646

• pytorch/torchtitan

  A PyTorch native platform for training generative AI models

  Language:Python4.4k56446511

• deepseek-ai/DualPipe

  A bidirectional pipeline parallelism algorithm for computation-communication overlap in V3/R1 training.

  Language:Python2.9k2911300

• HazyResearch/ThunderKittens

  Tile primitives for speedy kernels

  Language:Cuda2.7k4762174

• NVIDIA/cccl

  CUDA Core Compute Libraries

  Language:C++1.9k352.3k267

• tile-ai/tilelang

  Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

  Language:C++1.6k639154

• deepseek-ai/EPLB

  Expert Parallelism Load Balancer

  Language:Python1.3k2117195

• mit-han-lab/nunchaku

  [ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models

  Language:Cuda1.3k2224280

• deepseek-ai/profile-data

  Analyze computation-communication overlap in V3/R1.

  1.1k1716143

• zhihu/ZhiLight

  A highly optimized LLM inference acceleration engine for Llama and its variants.

  Language:C++9005217103

• alibaba/rtp-llm

  RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.

  Language:C++851129972

• aliyun/SimAI

  Language:C++6491286108

• vision-x-nyu/thinking-in-space

  Official repo and evaluation implementation of VSI-Bench

  Language:Python59641633

• geohot/cuda_ioctl_sniffer

  Sniff CUDA ioctls

  Language:C2076529

• bytedance/InfiniStore

  KV cache store for distributed LLM inference

  Language:C++13468313

• SubjectNoi/RTtention

  Language:Cuda20