--- title: "Open-source LLM fine-tuning and model training tools" type: "category" slug: "model-training" canonical_url: "https://www.graphcanon.com/categories/model-training" tool_count: 563 description: "Compare Unsloth, LlamaFactory, TRL, and Axolotl for fine-tuning. GraphCanon ranks open-source model training tools by GitHub stars and push freshness." --- # Model Training *GraphCanon updated Aug 21, 2026* Model training tools cover fine-tuning and post-training (LoRA, QLoRA, RLHF) when prompting and RAG are not enough for style, format, or domain behavior. Prefer lighter adapters before full fine-tunes; skip training entirely if retrieval already covers missing knowledge. GraphCanon lists published open-source toolkits with live GitHub signals, featured compares, and alternatives hubs for the highest-intent pairs. transformers currently leads this list at 164,121 GitHub stars among 563 published tools. 563 tools in this category (showing the top 60 by stars). ## Featured comparisons - [Unsloth vs Axolotl](/compare/axolotl-ai-cloud-axolotl-vs-unslothai-unsloth.md) - [Unsloth vs LlamaFactory](/compare/hiyouga-llamafactory-vs-unslothai-unsloth.md) - [Ollama vs Unsloth](/compare/ollama-ollama-vs-unslothai-unsloth.md) ## Alternatives hubs - [Unsloth alternatives](/tools/unslothai-unsloth/alternatives.md) - [LlamaFactory alternatives](/tools/hiyouga-llamafactory/alternatives.md) - [TRL alternatives](/tools/huggingface-trl/alternatives.md) ## Stacks - [The local / self-hosted LLM stack](/stacks/local-llm.md) ## Tools - [transformers](/tools/huggingface-transformers.md) - Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models (★ 164,121) [Very active] - [tensorflow](/tools/tensorflow-tensorflow.md) - An Open Source Machine Learning Framework for Everyone (★ 196,758) [Very active] - [LLMs-from-scratch](/tools/rasbt-llms-from-scratch.md) - Implement a ChatGPT-like LLM in PyTorch from scratch, step by step (★ 102,733) [Very active] - [keras](/tools/keras-team-keras.md) - Deep Learning for humans (★ 64,217) [Very active] - [LlamaFactory](/tools/hiyouga-llamafactory.md) - Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (★ 74,132) [Very active] - [unsloth](/tools/unslothai-unsloth.md) - A web UI for training and running open models locally. (★ 69,621) [Very active] - [JeecgBoot](/tools/jeecgboot-jeecgboot.md) - AI低代码平台，实现快速生成前后端系统及模块 (★ 47,405) [Very active] - [langflow](/tools/langflow-ai-langflow.md) - Tool for building and deploying AI-powered agents and workflows (★ 152,744) [Very active] - [ray](/tools/ray-project-ray.md) - Ray is an AI compute engine with a core distributed runtime and AI Libraries for accelerating ML workloads. (★ 43,526) [Very active] - [LibreChat](/tools/danny-avila-librechat.md) - Enhanced ChatGPT Clone with extensive features and integrations for self-hosting (★ 41,282) [Very active] - [mlflow](/tools/mlflow-mlflow.md) - AI engineering platform for debugging, evaluating, monitoring, and optimizing AI applications (★ 27,591) [Very active] - [netron](/tools/lutzroeder-netron.md) - Visualizer for neural network, deep learning and machine learning models (★ 33,302) [Very active] - [khoj](/tools/khoj-ai-khoj.md) - Your AI second brain. Self-hostable. (★ 36,512) [Active] - [RAG_Techniques](/tools/nirdiamant-rag-techniques.md) - Showcases advanced techniques for Retrieval-Augmented Generation (RAG) systems with detailed notebook tutorials. (★ 29,076) [Very active] - [peft](/tools/huggingface-peft.md) - State-of-the-art Parameter-Efficient Fine-Tuning (★ 21,443) [Very active] - [ai-guide](/tools/liyupi-ai-guide.md) - 免费开放的AI知识共享平台 (★ 18,766) [Active] - [ml-engineering](/tools/stas00-ml-engineering.md) - Machine Learning Engineering Open Book (★ 18,632) [Very active] - [generative-ai](/tools/googlecloudplatform-generative-ai.md) - Sample code and notebooks for Generative AI on Google Cloud, with Gemini Enterprise Agent Platform (★ 17,594) [Very active] - [bisheng](/tools/dataelement-bisheng.md) - BISHENG is an open LLM devops platform for next generation Enterprise AI applications (★ 11,879) [Very active] - [DeepSpeed](/tools/deepspeedai-deepspeed.md) - Deep learning optimization library for efficient distributed training and inference (★ 42,870) [Very active] - [ludwig](/tools/ludwig-ai-ludwig.md) - Low-code framework for building custom LLMs and AI models (★ 11,746) [Very active] - [mempalace](/tools/mempalace-mempalace.md) - The best-benchmarked open-source AI memory system. (★ 58,400) [Very active] - [OpenLLM](/tools/bentoml-openllm.md) - Run any open-source LLMs as OpenAI compatible API endpoint in the cloud. (★ 12,454) [Very active] - [pytorch](/tools/pytorch-pytorch.md) - Tensors and Dynamic neural networks in Python with strong GPU acceleration (★ 102,144) [Very active] - [unstructured](/tools/unstructured-io-unstructured.md) - Convert documents to structured data effortlessly (★ 15,238) [Very active] - [awesome-LLM-resources](/tools/wangrongsheng-awesome-llm-resources.md) - Summary of the world's best LLM resources. (★ 8,845) [Very active] - [oumi](/tools/oumi-ai-oumi.md) - Easily fine-tune, evaluate and deploy open source LLMs/VLMs (★ 9,359) [Very active] - [CosyVoice](/tools/funaudiollm-cosyvoice.md) - Multi-lingual large voice generation model with full-stack abilities for inference, training and deployment. (★ 22,373) [Steady] - [agent-lightning](/tools/microsoft-agent-lightning.md) - The absolute trainer to light up AI agents (★ 17,500) [Very active] - [train-llm-from-scratch](/tools/fareedkhan-dev-train-llm-from-scratch.md) - A straightforward method for training your LLM from raw text to aligned model generation (★ 9,141) [Very active] - [mcp-for-beginners](/tools/microsoft-mcp-for-beginners.md) - Open-source curriculum for Model Context Protocol (MCP), covering modular and secure AI workflows in multiple languages (★ 16,835) [Very active] - [BentoML](/tools/bentoml-bentoml.md) - The easiest way to serve AI apps and models (★ 8,793) [Active] - [awesome-gpt-image-2](/tools/youmind-openlab-awesome-gpt-image-2.md) - World's largest GPT Image 2 prompt library, updated daily (★ 8,852) [Very active] - [CV](/tools/accumulatemore-cv.md) - 超级全面的 深度学习 笔记 (★ 23,321) [Steady] - [metaflow](/tools/netflix-metaflow.md) - Build, Manage and Deploy AI/ML Systems (★ 10,228) [Very active] - [ART](/tools/openpipe-art.md) - Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. (★ 10,603) [Very active] - [Paddle](/tools/paddlepaddle-paddle.md) - High-performance single-machine and distributed deep learning & machine learning framework (★ 24,040) [Very active] - [llm-action](/tools/liguodongiot-llm-action.md) - Aims to share large model technology principles and practical experience (large model engineering, application implementation) (★ 24,898) [Active] - [garden-skills](/tools/conardli-garden-skills.md) - Skills collection for web design, knowledge retrieval, image generation and more. (★ 10,406) [Steady] - [tokenizers](/tools/huggingface-tokenizers.md) - 💥 Fast State-of-the-Art Tokenizers optimized for Research and Production (★ 10,940) [Very active] - [litgpt](/tools/lightning-ai-litgpt.md) - High-performance LLMs with recipes for pretraining, finetuning and deployment (★ 13,605) [Active] - [RD-Agent](/tools/microsoft-rd-agent.md) - Automating high-value R&D processes through AI-driven data science and model development. (★ 14,275) [Active] - [self-llm](/tools/datawhalechina-self-llm.md) - A guide for fine-tuning and deploying open-source large language models tailored for a Chinese audience on Linux. (★ 31,722) [Active] - [axolotl](/tools/axolotl-ai-cloud-axolotl.md) - Go ahead and axolotl questions (★ 12,297) [Very active] - [scikit-learn](/tools/scikit-learn-scikit-learn.md) - machine learning in Python (★ 66,855) [Very active] - [machine-learning-for-trading](/tools/stefan-jansen-machine-learning-for-trading.md) - Code for Machine Learning in Trading (★ 20,480) [Very active] - [wandb](/tools/wandb-wandb.md) - Weights & Biases platform for model training and management (★ 11,213) [Very active] - [langextract](/tools/google-langextract.md) - A Python library for extracting structured information from unstructured text using LLMs. (★ 38,400) [Very active] - [pytorch-lightning](/tools/lightning-ai-pytorch-lightning.md) - Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes. (★ 31,267) [Very active] - [juicefs](/tools/juicedata-juicefs.md) - Distributed POSIX file system built on Redis and S3 (★ 14,263) [Very active] - [LangGPT](/tools/langgptai-langgpt.md) - A pioneering framework for structured & meta-prompt design (★ 12,378) [Active] - [opendataloader-pdf](/tools/opendataloader-project-opendataloader-pdf.md) - PDF Parser for AI-ready data (★ 28,528) [Very active] - [llmfit](/tools/alexsjones-llmfit.md) - Hundreds of models & providers. One command to find what runs on your hardware. (★ 31,867) [Very active] - [llm-engineer-toolkit](/tools/kalyanks-nlp-llm-engineer-toolkit.md) - A curated list of over 120 LLM libraries categorized. (★ 10,767) [Very active] - [OpenRLHF](/tools/openrlhf-openrlhf.md) - Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (★ 9,891) [Active] - [skypilot](/tools/skypilot-org-skypilot.md) - Run, manage, and scale AI workloads on any AI infrastructure. (★ 10,456) [Very active] - [Megatron-LM](/tools/nvidia-megatron-lm.md) - Ongoing research training transformer models at scale (★ 17,341) [Very active] - [Speech](/tools/nvidia-nemo-speech.md) - A scalable generative AI framework for Speech AI (★ 17,940) [Very active] - [taipy](/tools/avaiga-taipy.md) - Turns Data and AI algorithms into production-ready web applications in no time. (★ 19,412) [Very active] - [amazon-sagemaker-examples](/tools/aws-amazon-sagemaker-examples.md) - Jupyter notebooks for building, training, and deploying ML models using Amazon SageMaker (★ 10,984) [Active] ## Common questions ### What are the best model training tools? GraphCanon ranks Model Training tools by GitHub adoption and freshness. transformers is the current leader (164,121 stars). See the full list on this page - sorted by stars, with [maintenance labels](/glossary/trust-and-signals/maintenance-label) and graph relationships. ### How does GraphCanon rank Model Training tools? We sort by GitHub stars and push recency on category pages, not paid placement. Alternatives and compare pages use [typed graph edges](/glossary/knowledge-graph/typed-edge) (alternative, successor, integrates_with) plus shared categories - constraint-first, not marketing votes. ### How many tools are in Model Training? 563 published tools are tagged with Model Training in the GraphCanon knowledge graph. ### What are popular Model Training comparisons? Head-to-head compare pages in this category include Unsloth vs Axolotl, Unsloth vs LlamaFactory, Ollama vs Unsloth. Each comparison uses live GitHub stats and optional [trust signals](/glossary/trust-and-signals/trust-signal) - see the comparisons block on this page. ### Which stacks use Model Training? Curated workflow pages that include Model Training: [The local / self-hosted LLM stack](/stacks/local-llm). Each stack step includes when-not-to-use guidance. ### Where are graph-backed alternatives hubs for Model Training? High-intent OSS-vs-OSS alternatives pages include [Unsloth alternatives](/tools/unslothai-unsloth/alternatives), [LlamaFactory alternatives](/tools/hiyouga-llamafactory/alternatives), [TRL alternatives](/tools/huggingface-trl/alternatives). Each hub ranks typed graph neighbors and constraint tags - not popularity votes. ### Is there a machine-readable Model Training list? Yes. Append `.md` to this URL or fetch [`/md/categories/model-training`](/md/categories/model-training) for a markdown twin. The JSON API exposes the same corpus at [`/api/graphcanon/categories/model-training`](/api/graphcanon/categories/model-training). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/categories/model-training`](/api/graphcanon/categories/model-training) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_