Efficient, Flexible, Multi-task Batch SFT Data Labeling Tool

Pretrain, finetune ANY AI model of ANY size on multiple GPUs, TPUs with zero code changes.

A modern Python-based MySQL backup tool with flexible archiving, multi-channel notifications (Telegram, Email, SMS, etc.), remote uploads (SFTP, FTP, SCP), and robust configuration validation.

Finetune Qwen3, Llama 4, TTS, DeepSeek-R1 & Gemma 3 LLMs 2x faster with 70% less memory! 🦥

sft training by using only command instruction on a ollama modelfile

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

This repo is making it accessible to fine tune your own LLM with SFT and DPO

A case study of a model evaluation and light fine-tuning. This features SFT, DAFT and a few PEFT methods.

25-2 YAI project

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Build SFT data from your PDFs

Generates high-quality fine-tuning pairs for large language models (LLMs) from unstructured documents.

This is a lightweight Decoder-Only language model with only 27MB of parameters, optimized for efficient text generation. It features multi-head attention, rotary embeddings, RMSNorm, and KV-cache, supporting fast inference and token generation.

Wrapped Megatron: As User-Friendly as HuggingFace, As Powerful as Megatron-LM | Megatron封装：和HuggingFace一样方便，和Megatron-LM一样强大

Training Qwen3 to solve Wordle using SFT and GRPO

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Upload folders faster via SFTP by temporarily zipping on the client and unzipping on the host.

A research sandbox for LLM pretraining, fine-tuning (SFT, DPO, RLHF), and alignment techniques. Designed for developing LLM-based solutions in computational biology.

An AI experiment - can an AI model teach itself to draw?

SFT on internlm2_1.8B_chat model