Instructions to use ddh0/phi-2-GGUF-fp16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ddh0/phi-2-GGUF-fp16 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="ddh0/phi-2-GGUF-fp16",
	filename="phi-2-fp16.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use ddh0/phi-2-GGUF-fp16 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ddh0/phi-2-GGUF-fp16
# Run inference directly in the terminal:
llama-cli -hf ddh0/phi-2-GGUF-fp16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ddh0/phi-2-GGUF-fp16
# Run inference directly in the terminal:
llama-cli -hf ddh0/phi-2-GGUF-fp16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf ddh0/phi-2-GGUF-fp16
# Run inference directly in the terminal:
./llama-cli -hf ddh0/phi-2-GGUF-fp16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf ddh0/phi-2-GGUF-fp16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ddh0/phi-2-GGUF-fp16

Use Docker

docker model run hf.co/ddh0/phi-2-GGUF-fp16

LM Studio
Jan

vLLM

How to use ddh0/phi-2-GGUF-fp16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ddh0/phi-2-GGUF-fp16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ddh0/phi-2-GGUF-fp16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ddh0/phi-2-GGUF-fp16

Ollama
How to use ddh0/phi-2-GGUF-fp16 with Ollama:
```
ollama run hf.co/ddh0/phi-2-GGUF-fp16
```

Unsloth Studio new

How to use ddh0/phi-2-GGUF-fp16 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 ddh0/phi-2-GGUF-fp16 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 ddh0/phi-2-GGUF-fp16 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ddh0/phi-2-GGUF-fp16 to start chatting

Docker Model Runner
How to use ddh0/phi-2-GGUF-fp16 with Docker Model Runner:
```
docker model run hf.co/ddh0/phi-2-GGUF-fp16
```

Lemonade

How to use ddh0/phi-2-GGUF-fp16 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ddh0/phi-2-GGUF-fp16

Run and chat with the model

lemonade run user.phi-2-GGUF-fp16-{{QUANT_TAG}}

List all available models

lemonade list

This is Microsoft's Phi-2, converted to GGUF without quantization. No other changes were made.

The model was converted using convert-hf-to-gguf.py from Georgi Gerganov's llama.cpp repo, release b1671.

All credit belongs to Microsoft for training and releasing this model. Thank you!

Downloads last month: 17

GGUF

Model size

3B params

Architecture

phi2

Hardware compatibility

We're not able to determine the quantization variants.

View all variants