Skip to content
Fossfits
Categories

Fine-tuning.

Train or adapt open models on your own GPU. LoRA, full fine-tunes, and tooling around datasets and recipes.

  1. Rank 1. UnslothFaster LoRA fine-tuning with lower VRAM on consumer GPUs.GitHub stars: 76.1k
  2. Rank 2. LLaMA-FactoryWeb UI and YAML recipes for fine-tuning many open LLMs.GitHub stars: 74.7k
  3. Rank 3. nanoGPTKarpathy’s ~300-line GPT trainer — the cleanest way to learn LLM training.GitHub stars: 63k
  4. Rank 4. AxolotlYAML-driven fine-tuning toolkit for serious training runs.GitHub stars: 12.5k
  5. Rank 5. DeepSpeedMicrosoft’s Apache-2.0 ZeRO optimizer for multi-GPU, multi-node LLM training.GitHub stars: 43.1k
  6. Rank 6. TRLHugging Face’s Apache-2.0 trainers for SFT, DPO, PPO, and reward modeling.GitHub stars: 19.3k
  7. Rank 7. torchtunePyTorch-native library for fine-tuning LLMs with recipes.GitHub stars: 5.8k
  8. Rank 8. PEFTHugging Face’s LoRA/QLoRA library that makes big-model tuning fit one GPU.GitHub stars: 21.7k