Finetune Llama 3.3, Mistral, Phi-4, Qwen 2.5 & Gemma 2-5x faster with 80% less memory! All notebooks are beginner friendly! Add your dataset, click Run All, and youll get a 2x faster finetuned model which can be exported to GGUF, Ollama, vLLM or uploaded to Hugging Face. Unsloth supports Llama 3.2 (3B), Phi-4 (14B), Llama 3.2 Vision (11B), Llama 3.1 (8B), Gemma 2 (9B), Qwen 2.5 (7B), Mistral v0.3 (7B), Ollama, ORPO, DPO Zephyr. See all their notebooks and models on the Unsloth website. Kaggle Notebooks available for Llama 3.2, Llama 3.1 (8B), Gemma 2 (9B), and Mistral (7B). Run notebooks for Llama 3.2 conversational, Llama 3.1 conversational, and Mistral v0.3 ChatML. Also check out the text completion notebook for continued pretraining / raw text.
