# Train LLM from Scratch alternatives

Train LLM from Scratch: A straightforward method for training your llm from scratch.

24 products that do a similar job, closest first. 11 have pricing checked on their own site; 10 of those are free or have a free tier.

Page: https://nowfound.app/alternatives/train-llm-from-scratch
About Train LLM from Scratch: https://nowfound.app/launch/train-llm-from-scratch.md

- [Lathe](https://nowfound.app/launch/lathe-use-llms-to-learn-a-new-domain-not-skip-past-it): Lathe is a learning tool that generates hands-on tutorials for technical topics using LLMs, then guides users through building projects themselves via a local web interface. (2026-06-07, 402 votes, free, github.com)
- [guppylm](https://nowfound.app/launch/i-built-a-tiny-llm-to-demystify-how-language-models-work): Built a ~9M param LLM from scratch to understand how they actually work. Vanilla transformer, 60K synthetic conversations, ~130 lines of PyTorch. Trains in 5 min on a free Colab T4. The fish thinks t… (2026-04-06, 915 votes, free, github.com)
- [Create-LLM](https://nowfound.app/launch/create-llm-train-your-own-llm-in-60-seconds): Create-LLM is a tool that enables users to train their own large language models in approximately 60 seconds. (2025-10-26, 54 votes, free, github.com)
- [Supervised app](https://nowfound.app/launch/supervised-app): Build LLMs powered by GPT & your own data (2023-06-23, 238 votes)
- [ML From Scratch](https://nowfound.app/launch/ml-from-scratch-free-online-textbook): ML From Scratch is a free online textbook covering machine learning fundamentals. (2020-08-31, 259 votes, dafriedman97.github.io)
- [Building an LLM from Scratch in TypeScript](https://nowfound.app/launch/building-an-llm-from-scratch-in-typescript): The from-scratch guide to building large language models in NodeJS. (2026-10-14, 33 votes, book.humanexperiences.io)
- [Llms.txt Generator](https://nowfound.app/launch/llms-txt-generator-st651034): Generate a free llms.txt file for your website that helps LLMs discover and use your website and improve your website's AI SEO. (2024-11-27, 145 votes, free tier, from $29/mo, sitespeak.ai)
- [Learn LLMs LeetCode Style](https://nowfound.app/launch/learn-llms-leetcode-style): LeetCode for PyTorch — 65 ML/AI interview problems from real interviews at Google, Meta, Anthropic. Jupyter notebooks, an auto-grader, and an MCP AI tutor. (2025-07-13, 181 votes, free, github.com)
- [UpTrain](https://nowfound.app/launch/uptrain-a-practical-approach-to-finetuning-llms-for-custom-u): UpTrain is an open-source framework designed to help developers finetune large language models for specific applications. (2023-03-30, 10 votes, github.com)
- [How LLMs Work](https://nowfound.app/launch/how-llms-work-interactive-visual-guide-based-on-karpathy-s-l): This is an interactive visual guide explaining how large language models work, based on Andrej Karpathy's "Intro to Large Language Models" lecture. (2026-04-24, 245 votes, ynarwal.github.io)
- [Train a language model to talk like you](https://nowfound.app/launch/train-a-language-model-to-talk-like-you): This tool lets users train a language model to replicate their writing style or speech patterns. (2020-01-21, 209 votes, colab.research.google.com)
- [Local LLMs by Sttabot AI](https://nowfound.app/launch/local-llms-by-sttabot-ai): Build local LLMs using top data science libraries (2023-08-12, 112 votes)
- [LLM Scraper](https://nowfound.app/launch/llm-scraper-turn-any-webpage-into-structured-data): Turn any webpage into structured data using LLMs. (2024-04-20, 88 votes, free, github.com)
- [Recursive LLM Prompts](https://nowfound.app/launch/recursive-llm-prompts): Recursive LLM Prompts is a tool that uses language models to implement recursion through English-language prompts rather than traditional code. (2023-03-20, 97 votes, free, github.com)
- [Dolly](https://nowfound.app/launch/dolly): Democratizing the magic of ChatGPT with open models (2023-03-27, 113 votes, free, github.com)
- [Map with an LLM](https://nowfound.app/launch/map-with-an-llm): Map with an LLM is an AI tool that helps users identify ideal locations by combining mapping capabilities with large language model technology. (2024-12-27, 76 votes, godview.ai)
- [Slate](https://nowfound.app/launch/slate-st446457): A playground for ideas - LLM prompts in code notebooks (2024-03-25, 93 votes)
- [Data Science from Scratch](https://nowfound.app/launch/data-science-from-scratch): Data Science and Python, starting at zero (2015-11-23, 76 votes, joelgrus.com)
- [GradientJ](https://nowfound.app/launch/gradientj): Build complex LLM applications fast and manage them at scale (2023-05-24, 92 votes, tryvelos.com)
- [Dataset Labs](https://nowfound.app/launch/dataset-labs): The fastest way to build LLM training data (2026-01-11, 1 votes, free tier, datasetlabs.ai)
- [LLMWare](https://nowfound.app/launch/llmware-small-specialized-function-calling-1b-llms-for-multi): LLMWare is an open-source RAG framework featuring specialized 1B language models designed for multi-step retrieval-augmented generation workflows. (2024-02-11, 51 votes, free, github.com)
- [CookLLM](https://nowfound.app/launch/cookllm): Go deep. Build real things. (2026-02-26, 2 votes, one-time $69, cookllm.com)
- [TEXT LLM](https://nowfound.app/launch/text-llm): All Your Repository code for your LLM (2025-04-22, 12 votes, textllm.vercel.app)
- [Trained Tiny Tales GPT(30M model)from scratch](https://nowfound.app/launch/trained-tiny-tales-gpt-30m-model-from-scratch-and-deployed-i): Trained Tiny Tales GPT is a 30-million parameter language model trained from scratch on 1 billion tokens and deployed as a web service. (2024-04-17, 18 votes, kunalmishra.info)
