See our collection for versions of Phi 4 including GGUF, 4 bit & more formats. Dynamic 4 bit: Unsloth's Dynamic 4 bit Quants selectively avoids quantizing certain parameters, greatly increase accuracy than standard 4 bit. See our full collection of Unsloth quants on Hugging Face here. Finetune Phi 4, Llama 3.3 2 5x faster with 70% less memory via Unsloth! We have a free Google Colab notebook for Phi 4 here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Phi 4 Conversational.ipynb unsloth/phi 4 unsloth bnb 4bit We have converted Phi 4 to Llama's architecture for improved ease of use, better fine tuning, and greater accuracy. Also contains Unsloth's Phi 4 bugfixes. ✨ Finetune for Free All notebooks are beginner friendly ! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model. Unsloth supports Free Notebooks Performance Memory use Phi 4 ▶️ Start on Colab 2x faster 50% less Llama 3.2 (3B) ▶️ Start on Colab Conversational.ipynb) 2.4x faster 58% less Llama 3.2 (11B vision) ▶️ Start on Colab Vision.ipynb) 2x faster 60% less Qwen2 VL (7B) ▶️ Start on Colab Vision.ipynb) 1.8x faster 60% less Qwen2.5 (7B) ▶️ Start on Colab Alpaca.ipynb) 2x f…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy