by FRS2003 · LLM Plugin · ★ 79
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
动手学大模型全栈：CS336 中文精讲 · PyTorch 手搓 Transformer · 单卡复现 Pretrain/SFT/LoRA/DPO/GRPO/RLVR · PEFT/Agent/RAG 落地（含真实实验、曲线与复现脚本）
| Stars | 79 |
| Forks | 10 |
| Language | Python |
| Category | LLM Plugin |
| License | MIT |
| Quality Score | 53.7711684947593/100 |
| Last Updated | 2026-09-16 |
| Created | 2026-06-25 |
| Platforms | python |
| Est. Tokens | ~13k |
Looking for a hands-on-llm alternative? If you're comparing hands-on-llm with other llm plugin tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Zero-friction LLM fine-tuning skill for Claude Code, Gemini CLI & any ACP agent. Unsloth on NVIDIA · TRL+MPS/M
🦀 Decoder-only LLM built from scratch in pure Rust using Candle — no Python, no PyTorch. Gated DeltaNet + spa
A travel agent based on Qwen2.5, fine-tuned by SFT + DPO/PPO/GRPO using traveling question-answer dataset, a m
The fastest way to build and start training your own LLM. CLI tool that scaffolds production-ready PyTorch tra
Awesome LLM Papers and repos on very comprehensive topics.
🌱 A little course on Reinforcement Learning Environments for evaluating and training Language Models
Explore other popular llm plugin tools:
hands-on-llm is 动手学大模型全栈：CS336 中文精讲 · PyTorch 手搓 Transformer · 单卡复现 Pretrain/SFT/LoRA/DPO/GRPO/RLVR · PEFT/Agent/RAG 落地（含真实实验、曲线与复现脚本）. It is categorized as a LLM Plugin with 79 GitHub stars.
hands-on-llm is primarily written in Python. It covers topics such as cs336, deep-learning, dpo.
You can find installation instructions and usage details in the hands-on-llm GitHub repository at github.com/FRS2003/hands-on-llm. The project has 79 stars and 10 forks, indicating an active community.
hands-on-llm is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to hands-on-llm on Agent Skills Hub include unsloth-buddy, aarambh-studio, Travel-Agent-based-on-Qwen2-RLHF. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: