Stars
Finetune Llama 3.3, DeepSeek-R1 & Reasoning LLMs 2x faster with 70% less memory! 🦥
Awesome coreset/core-set/subset/sample selection works.
Fully open reproduction of DeepSeek-R1
This repository collects papers for "A Survey on Knowledge Distillation of Large Language Models". We break down KD into Knowledge Elicitation and Distillation Algorithms, and explore the Skill & V…
A modular graph-based Retrieval-Augmented Generation (RAG) system
Fast and flexible image augmentation library. Paper about the library: https://www.mdpi.com/2078-2489/11/2/125
Exclusively Dark (ExDARK) dataset which to the best of our knowledge, is the largest collection of low-light images taken in very low-light environments to twilight (i.e 10 different conditions) to…
仅需Python基础,从0构建大语言模型;从0逐步构建GLM4\Llama3\RWKV6, 深入理解大模型原理
从无名小卒到大模型(LLM)大英雄~ 欢迎关注后续!!!
Janus-Series: Unified Multimodal Understanding and Generation Models
Ongoing research training transformer models at scale
SGLang is a fast serving framework for large language models and vision language models.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Get up and running with Llama 3.3, DeepSeek-R1, Phi-4, Gemma 2, and other large language models.
PromptInject is a framework that assembles prompts in a modular fashion to provide a quantitative analysis of the robustness of LLMs to adversarial prompt attacks. 🏆 Best Paper Awards @ NeurIPS ML …
Data validation using Python type hints
NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.
LLMServingSim: A HW/SW Co-Simulation Infrastructure for LLM Inference Serving at Scale
A lightweight, low-dependency, unified API to use all common reranking and cross-encoder models.
Fork of https://huggingface.co/hexgrad/Kokoro-82M
SoftVC VITS Singing Voice Conversion
1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
Official Implementation of CVPR24 highligt paper: Matching Anything by Segmenting Anything
[CVPR 2023] DepGraph: Towards Any Structural Pruning
Code release for "Detecting Twenty-thousand Classes using Image-level Supervision".