lora

pytorch를 사용하여 텍스트 전처리부터 RAG, 에이전트, LLM 파인튜닝을 정리한 Deep Learning NLP 저장소입니다.

Fine-tuning of diffusion models

Build your own offline AI from any documents. Free. No coding. LoRA fine-tuning + RAG + GGUF export.

FlowSteer: agents designing agentic workflows via reinforced progressive canvas editing.

🚀 Easy, open-source LLM finetuning with one-line commands, seamless cloud integration, and popular optimization frameworks. ✨

MCore-Bridge: Providing Megatron-Core model definitions for state-of-the-art large models and making Megatron training as simple as Transformers — with support for 300+ large language models (Qwen3-Next, GLM-5.2, Deepseek-V4, MiniMax-2.7, ...) and 200+ multimodal large models (Qwen3.5, Qwen3-Omni, Gemma4, ...).

Finetune mistral-7b-instruct for sentence embeddings

Spirit Lora Trainer is a robust toolkit for training Flux1-LoRA models with a focus on simplicity and reliability and based on kohya-ss script. 智灵训练器.

Official PyTorch Implementation for the "Recovering the Pre-Fine-Tuning Weights of Generative Models" paper (ICML 2024).

A Diffusion training toolbox based on diffusers and existing SOTA methods, including Dreambooth, Texual Inversion, LoRA, Custom Diffusion, XTI, ....

A travel agent based on Qwen2.5, fine-tuned by SFT + DPO/PPO/GRPO using traveling question-answer dataset, a mindmap can be output using the response. A RAG system is build upon the tuned qwen2, using Prompt-Template + Tool-Use + Chroma embedding database + LangChain

diffusion lora chinese tutorial，虚拟idol训练中文教程

PyTorch Reimplementation of LoRA (featuring with supporting nn.MultiheadAttention in OpenCLIP)

This repo contains a list of channels and sources from where LLMs should be learned

🌹 Rose: Range-Of-Slice Equilibration PyTorch optimizer. Stateless optimization through range-normalized gradient updates.

LoRA Pilot is an ultimate docker image for all Stable Diffusion LoRA trainers. Includes kohya_ss, diffusion pipes and TensorBoard for trainings and ComfyUI and InvokeAI for validation. Features shared models, modules, custom integrations and automatization scripts.

Emotion text classification using Llama3-8b with LoRA and FlashAttention. Based on LLaMA-Factory.

Real-time football video analytics: player/ball/referee tracking and pitch keypoints. YOLO, HRNet, OSNet, ByteTrack, FastAPI. Game state recognition with lightweight CLIP/VLM validation. Python, OpenCV, CUDA/MPS.

Distill teacher chains-of-thought into a LoRA adapter via a strict boxed-answer format contract + two-phase Train→Nudge (silver-medal NVIDIA Nemotron reasoning recipe, as a tested library).

LoRa trainer for SDXL and SD15