Ma-Dan

calm

CUDA/Metal accelerated language model inference

infini-mini-transformer

This is a personal reimplementation of Google's Infini-transformer, utilizing a small 2b model. The project includes both model and training code.

gemma-sft

Gemma-SFT, gemma-2b/gemma-7b微调(finetune,transformers)/LORA(peft)/推理(inference)

vtensor

text-generation-inference

Large Language Model Text Generation Inference

exllamav2

A fast inference library for running LLMs locally on modern consumer-class GPUs

json

JSON for Modern C++

llamafile

Distribute and run LLMs with a single file.

chinese-independent-developer

👩🏿‍💻👨🏾‍💻👩🏼‍💻👨🏽‍💻👩🏻‍💻**独立开发者项目列表 -- 分享大家都在做什么

ruapu

Detect CPU features with single-file

llm.c

LLM training in simple, raw C/CUDA

Qwen1.5-0.5b-chat-android

:fire: 基于MNN-llm的安卓手机部署大语言模型：Qwen1.5-0.5B-Chat

lut-gemm

GitHub-Chinese-Top-Charts

:cn: GitHub中文排行榜，各语言分设「软件 | 资料」榜单，精准定位中文好项目。各取所需，高效学习。

llvm-project

The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.

CUDA_gemm

A simple high performance CUDA GEMM implementation.

RAGOnMedicalKG

RAGOnMedicalKG，将大模型RAG与KG结合，完成demo级问答，旨在给出基础的思路。

mamba-minimal

Simple, minimal implementation of the Mamba SSM in one file of PyTorch.

faster-whisper

Faster Whisper transcription with CTranslate2

verilog_learning

RasaGPT

💬 RasaGPT is the first headless LLM chatbot platform built on top of Rasa and Langchain. Built w/ Rasa, FastAPI, Langchain, LlamaIndex, SQLModel, pgvector, ngrok, telegram

streaming-tts-webui

Streaming Text to Speech Web UI

sqlflow

Brings SQL and AI together.

insanely-fast-whisper

flash_attn_jax

JAX bindings for Flash Attention v2

CUDA-Learn-Notes

🎉CUDA 笔记 / 大模型手撕CUDA / C++笔记，更新随缘: flash_attn、sgemm、sgemv、warp reduce、block reduce、dot product、elementwise、softmax、layernorm、rmsnorm、hist etc.

diffusion_schrodinger_bridge

PyTorch Implementation of DSB for Score Based Generative Modeling. Experiments managed using Hydra.

euanka

本地完整部署ASR(K2)-NLP(Rasa,Spacy)-LLM(Chatglm2)-TTS(Vits)

rwkv-qualcomm

Inference rwkv5 with Qualcomm AI Engine Direct SDK

AllNewsSpider

澎湃新闻，新浪新闻，腾讯新闻，搜狐新闻，新闻联播，泰晤士报，纽约时报，BBCNews，旨在爬取所有新闻门户网站的新闻，禁止将所得数据商用！