Library
85,340 resources from 11 public sources. 54 checked by hand; the rest are unreviewed.
Results
-
verl-agent
Agents & frameworksverl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Policy Optimization for LLM Agent Training"
git clone https://github.com/langfengQ/verl-agent.git -
Nemotron-Post-Training-Dataset-v1
Datasets & benchmarksHugging Face dataset by nvidia
nvidia/Nemotron-Post-Training-Dataset-v1 -
AgileRL
Agents & frameworksThe complete post-training stack for LLMs, plus RL for any problem. SFT, DPO and multi-turn agentic RL, with evolutionary hyperparameter optimization built in.
git clone https://github.com/AgileRL/AgileRL.git -
agent-lightning
AI repositoriesThe absolute trainer to light up AI agents.
git clone https://github.com/microsoft/agent-lightning.git -
train_1M_CN
Datasets & benchmarksHugging Face dataset by BelleGroup
BelleGroup/train_1M_CN -
train_3.5M_CN
Datasets & benchmarksHugging Face dataset by BelleGroup
BelleGroup/train_3.5M_CN -
zephyr-7b-beta
Modelstext generation model by HuggingFaceH4 · transformers
HuggingFaceH4/zephyr-7b-beta -
Libra
Agents & frameworksLibra: Efficient Resource Management for Agentic RL Post-Training
git clone https://github.com/NetX-lab/Libra.git -
stack-v3-train
Datasets & benchmarksHugging Face dataset by HuggingFaceCode
HuggingFaceCode/stack-v3-train -
zephyr-7b-alpha
Modelstext generation model by HuggingFaceH4 · transformers
HuggingFaceH4/zephyr-7b-alpha -
12306-mcp
MCP serversThis is a 12306 ticket search server based on the Model Context Protocol (MCP).
git clone https://github.com/Joooook/12306-mcp.git -
SwanLab
AI repositories⚡️SwanLab - an open-source, modern-design AI training tracking and visualization tool. Supports Cloud / Self-hosted use. Integrated with PyTorch / Transformers / verl / LLaMA Factory / ms-swift / Ultralytics / MMEngine / Keras etc.
git clone https://github.com/SwanHubX/SwanLab.git -
metaflow
AI repositoriesBuild, Manage and Deploy AI/ML Systems
git clone https://github.com/Netflix/metaflow.git -
OpenVid-1M
Datasets & benchmarksHugging Face dataset by nkp37
nkp37/OpenVid-1M -
heretic
AI repositoriesTraining-free language model modification: Remove censorship, suppress slop, reduce sycophancy, and more
git clone https://github.com/p-e-w/heretic.git -
data-juicer
AI repositoriesData processing for and with foundation models! 🍎 🍋 🌽 ➡️ ➡️🍸 🍹 🍷
git clone https://github.com/datajuicer/data-juicer.git -
llm-datasets
AI repositoriesCurated list of datasets and tools for post-training.
git clone https://github.com/mlabonne/llm-datasets.git -
llm-algo-leetcode
Docs & tutorialsHands-on LLM algorithms and systems with PyTorch, Triton & CUDA.《大模型算法与系统教程》:模型架构与训练、后训练、推理加速、性能与算子优化,配套 Notebook 练习、答案解析、测试与实验。
git clone https://github.com/datawhalechina/llm-algo-leetcode.git -
dstack
SkillsA unified orchestration layer for heterogeneous AI compute. It standardizes how to manage compute and run training and inference on GPU clouds, Kubernetes, VMs, or bare-metal clusters.
git clone https://github.com/dstackai/dstack.git -
PaddleNLP
AI repositoriesEasy-to-use and powerful LLM and SLM library with awesome model zoo.
git clone https://github.com/PaddlePaddle/PaddleNLP.git -
awsome-distributed-ai
AI repositoriesBest practices, reference architectures, and examples for distributed AI training and inference on AWS.
git clone https://github.com/awslabs/awsome-distributed-ai.git -
Nemotron-Post-Training-Dataset-v2
Datasets & benchmarksHugging Face dataset by nvidia
nvidia/Nemotron-Post-Training-Dataset-v2 -
remote-gpu-trainer
SkillsAn Agent Skill for Claude Code, Codex and other coding agents: train on rented GPUs without losing checkpoints or paying for idle machines.
git clone https://github.com/Hanyuyuan6/remote-gpu-trainer.git -
doc-adk-training
Docs & tutorialsLearn to build, test, and deploy production-ready AI agents with this complete "From Zero to Hero" training course for the Google Agent Development Kit (ADK).
git clone https://github.com/mauripsale/doc-adk-training.git
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.