Instructions to use cortexso/phi-4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cortexso/phi-4 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="cortexso/phi-4",
	filename="phi-4-q2_k.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use cortexso/phi-4 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cortexso/phi-4:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/phi-4:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cortexso/phi-4:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/phi-4:Q4_K_M

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 cortexso/phi-4:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf cortexso/phi-4:Q4_K_M

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 cortexso/phi-4:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf cortexso/phi-4:Q4_K_M

Use Docker

docker model run hf.co/cortexso/phi-4:Q4_K_M

LM Studio
Jan

vLLM

How to use cortexso/phi-4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cortexso/phi-4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cortexso/phi-4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cortexso/phi-4:Q4_K_M

Ollama
How to use cortexso/phi-4 with Ollama:
```
ollama run hf.co/cortexso/phi-4:Q4_K_M
```

Unsloth Studio new

How to use cortexso/phi-4 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 cortexso/phi-4 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 cortexso/phi-4 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for cortexso/phi-4 to start chatting

Docker Model Runner
How to use cortexso/phi-4 with Docker Model Runner:
```
docker model run hf.co/cortexso/phi-4:Q4_K_M
```

Lemonade

How to use cortexso/phi-4 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull cortexso/phi-4:Q4_K_M

Run and chat with the model

lemonade run user.phi-4-Q4_K_M

List all available models

lemonade list

Overview

Phi-4 model, a state-of-the-art 14B parameter Transformer designed for advanced reasoning, conversational AI, and high-quality text generation. Built on a mix of synthetic datasets, filtered public domain content, academic books, and Q&A datasets, Phi-4 ensures exceptional performance through data quality and alignment. It features a 16K token context length, trained on 9.8T tokens over 21 days using 1920 H100-80G GPUs. Phi-4 underwent rigorous fine-tuning and preference optimization to enhance instruction adherence and safety. Released on December 12, 2024, it represents a static model with data cutoff as of June 2024, suitable for diverse applications in research and dialogue systems.

Variants

No	Variant	Cortex CLI command
1	Phi-4-14b	`cortex run phi-4:14b`

Use it with Jan (UI)

Install Jan using Quickstart
Use in Jan model Hub:
```
cortexso/phi-4
```

Use it with Cortex (CLI)

Install Cortex using Quickstart
Run the model with command:
```
cortex run phi-4
```

Credits

Author: Microsoft Research
Converter: Homebrew
Original License: License
Papers: Phi-4 Technical Report

Downloads last month: 143

GGUF

Model size

15B params

Architecture

phi3

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Paper for cortexso/phi-4

Phi-4 Technical Report

Paper • 2412.08905 • Published Dec 12, 2024 • 123