A high-throughput and memory-efficient inference and serving engine for LLMs

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

所有小初高、大学PDF教材。

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.

Fast and memory-efficient exact attention

verl: Volcano Engine Reinforcement Learning for LLMs

A book for Learning the Foundations of LLMs

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3, Qwen3-MoE, DeepSeek-R1, GLM4.5, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Llava, …

OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.

MedicalGPT: Training Your Own Medical GPT Model with ChatGPT Training Pipeline. 训练医疗大模型，实现了包括增量预训练(PT)、有监督微调(SFT)、RLHF、DPO、ORPO、GRPO。

Sky-T1: Train your own O1 preview model within $450

Code for the paper "Evaluating Large Language Models Trained on Code"

A streamlined and customizable framework for efficient large model (LLM, VLM, AIGC) evaluation and performance benchmarking.

中文法律LLaMA (LLaMA for Chinese legel domain)

Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (LLM).

Official Repository of "Learning to Reason under Off-Policy Guidance"

PyContinual (An Easy and Extendible Framework for Continual Learning)

Official repo of Toucan: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments

[ICLR 2025] Released code for paper "Spurious Forgetting in Continual Learning of Language Models"

Continual Learning for Transformers that allows training on multiple tasks sequentially while preserving knowledge from earlier tasks using Elastic Weight Consolidation.

Source code for a LoRA-based continual relation extraction method.