Instructions to use unsloth/gemma-4-E4B-it-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Local Apps Settings
- Unsloth Studio
How to use unsloth/gemma-4-E4B-it-NVFP4 with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for unsloth/gemma-4-E4B-it-NVFP4 to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for unsloth/gemma-4-E4B-it-NVFP4 to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for unsloth/gemma-4-E4B-it-NVFP4 to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="unsloth/gemma-4-E4B-it-NVFP4", max_seq_length=2048, )
Updated all with official Gemma chat template update
#1
by danielhanchen - opened
17th July 2026: We updated all uploads to include Gemma's new official chat template improvements which increased tool call accuracy by up to 10% on some benchmarks! See https://x.com/googlegemma/status/2077449152062247219 for more details. Please re-download and or use --chat-template-file for llama.cpp.