karpathy/nanoGPT

Code for training and fine-tuning GPT language models

View on index · View in 3D Map

Related repos

• competes to train GPT-2 models super fast on 8xH100 GPUs using speedrun benchmarks
• train a ChatGPT-like model from scratch for $100 in 4 hours
• Code and tutorials for building GPT-style language models from the ground up in PyTorch
• litgpt: tools to train, finetune, and run large language models from the command line
• minimal PyTorch code for training your own small language model from scratch
• AI tool that turns text instructions into code projects automatically
• Streaming speech-to-speech language model
• End-to-end multimodal LLM for any-to-any I/O
• migrate codebases between languages or frameworks using AI
• open-weight GPT models for reasoning and agent tasks