@dailydoseofds_ · Daily Dose of Data Science
Saved 2026-05-15 · Posted 2026-03-12 · Status: New
Fine-tune 100+ LLMs directly from a UI without any code! 💻
(100% open-source with 68K+ stars)
LLaMA-Factory lets you train and fine-tune open-source LLMs and VLMs without writing any code.
It supports:
✅ Almost all popular models (LLaMA, Mistral, DeepSeek, Gemma, etc.)
✅ Efficient fine-tuning (LoRA, QLoRA, DoRA, LoRA+, etc.)
✅ Integrated methods (PPO, DPO, KTO, ORPO, etc.)
✅ Practical tricks (Flash Attention, RoPE scaling, etc.)
✅ Experiment monitoring (TensorBoard, W&B, MLflow, etc.)
✅ Several downstream tasks (Tool use, multimodal understanding, etc.)
Perfect for researchers and engineers who want to experiment with fine-tuning without diving into code complexity.
GitHub repo in the comments!
👉 Over to you: Have you fine-tuned any LLMs yet?
#ai #llm #finetuning
Content ideas (0)
No ideas generated yet. Run /instagram-sync ideate from Claude Code to create some.
Comments (7)
GitHub Repo → https://github.com/hiyouga/LLaMA-Factory
WWOOWW WOWOW holds man holds onnss man that cant be fair, hells no so u tellin me we can easily finetune an ai to be a lawyer, pharamnsist, and stuff liek that so easy?
Plz dont give me hope
😮
Beautiful
🧑🏿💻📝✍️🏿