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

Libraries

How to use microsoft/phi-4-gguf with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="microsoft/phi-4-gguf")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("microsoft/phi-4-gguf", dtype="auto")

llama-cpp-python

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

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="microsoft/phi-4-gguf",
	filename="phi-4-IQ3_M.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 microsoft/phi-4-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf microsoft/phi-4-gguf:Q4_K_S
# Run inference directly in the terminal:
llama-cli -hf microsoft/phi-4-gguf:Q4_K_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf microsoft/phi-4-gguf:Q4_K_S
# Run inference directly in the terminal:
llama-cli -hf microsoft/phi-4-gguf:Q4_K_S

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 microsoft/phi-4-gguf:Q4_K_S
# Run inference directly in the terminal:
./llama-cli -hf microsoft/phi-4-gguf:Q4_K_S

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 microsoft/phi-4-gguf:Q4_K_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf microsoft/phi-4-gguf:Q4_K_S

Use Docker

docker model run hf.co/microsoft/phi-4-gguf:Q4_K_S

LM Studio
Jan

vLLM

How to use microsoft/phi-4-gguf with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/microsoft/phi-4-gguf:Q4_K_S

SGLang

How to use microsoft/phi-4-gguf with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "microsoft/phi-4-gguf" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/phi-4-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "microsoft/phi-4-gguf" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/phi-4-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use microsoft/phi-4-gguf with Ollama:
```
ollama run hf.co/microsoft/phi-4-gguf:Q4_K_S
```

Unsloth Studio new

How to use microsoft/phi-4-gguf 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 microsoft/phi-4-gguf 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 microsoft/phi-4-gguf to start chatting

Using HuggingFace Spaces for Unsloth

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

Docker Model Runner
How to use microsoft/phi-4-gguf with Docker Model Runner:
```
docker model run hf.co/microsoft/phi-4-gguf:Q4_K_S
```

Lemonade

How to use microsoft/phi-4-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull microsoft/phi-4-gguf:Q4_K_S

Run and chat with the model

lemonade run user.phi-4-gguf-Q4_K_S

List all available models

lemonade list

Gustavo de Rosa commited on May 8, 2025

Commit

18ece48

1 Parent(s): b1e764c

chore(root): Re-convert and re-quantize files.

Browse files

Files changed (24) hide show

phi-4-q4.gguf → phi-4-IQ2_M.gguf +2 -2
phi-4-fp16.gguf → phi-4-IQ3_M.gguf +2 -2
phi-4-IQ3_S.gguf +3 -0
phi-4-IQ3_XS.gguf +3 -0
phi-4-IQ3_XXS.gguf +3 -0
phi-4-IQ4_NL.gguf +3 -0
phi-4-IQ4_XS.gguf +3 -0
phi-4-Q2_K.gguf +3 -0
phi-4-Q3_K.gguf +3 -0
phi-4-Q3_K_L.gguf +3 -0
phi-4-Q3_K_S.gguf +3 -0
phi-4-Q4_0.gguf +3 -0
phi-4-Q4_1.gguf +3 -0
phi-4-Q4_K.gguf +3 -0
phi-4-Q4_K_S.gguf +3 -0
phi-4-Q5_0.gguf +3 -0
phi-4-Q5_1.gguf +3 -0
phi-4-Q5_K.gguf +3 -0
phi-4-Q5_K_S.gguf +3 -0
phi-4-Q6_K.gguf +3 -0
phi-4-Q8_0.gguf +3 -0
phi-4-TQ1_0.gguf +3 -0
phi-4-TQ2_0.gguf +3 -0
phi-4-bf16.gguf +3 -0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab704ffa097f5902b937ca5d0c166226f1201e7492da30fa4e1c086c217afd6b
-size 9053114464

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eefbdf36370dcac18aa8d15bace19e30be3a6e221e5448b57f7fa27ea197f2eb
-size 29323399264

 version https://git-lfs.github.com/spec/v1
+oid sha256:a3173c8a4f826e443866bd750bf1acebd4bee915dd74c6aa635a1c74e4c1d850
+size 588873920

 version https://git-lfs.github.com/spec/v1
+oid sha256:c9298fa6c9e927b257845a8057c74f2f4bd6bb7fbce35a0db933debdd777ae9a
+size 6913835200

+version https://git-lfs.github.com/spec/v1
+oid sha256:3f165ed951be668c9297e0721dc8a8fdab4916a8da7a6b83a3b47a2a174a9c0e
+size 6504747200

+version https://git-lfs.github.com/spec/v1
+oid sha256:2bd63e342a7d5801f45cab2fb15522812e9c1f4731f92b5a349474743876121c
+size 6246699200

+version https://git-lfs.github.com/spec/v1
+oid sha256:b65f01aab058dc26f8f1d0a4fe84b68df58d29ea2f21c87972620a9869e57bc5
+size 6079643840

+version https://git-lfs.github.com/spec/v1
+oid sha256:53aa11369a090f667dabf6b5877baa9f9947071bf7e7244a737b2d7b1d5c0379
+size 8440762560

+version https://git-lfs.github.com/spec/v1
+oid sha256:f5c151b6e5ea5e5bfd699b40e4c6797d08a4c6c0136cad201fb03e1d3e6e5511
+size 8013058240

+version https://git-lfs.github.com/spec/v1
+oid sha256:7ff048b61e9d6fc49bdd11affdb5e0d3365c08e55eb17ffd0f159f6e0de6f53e
+size 5547348160

+version https://git-lfs.github.com/spec/v1
+oid sha256:570a44b43fd7bd87637586ac8d7fc47f5a2fde0376d4f58c22854960c43eb365
+size 7363268800

+version https://git-lfs.github.com/spec/v1
+oid sha256:8fe8deaa431f877d7c9c3caf49f0e149c396b526c50558991ad87040cd43111e
+size 7930155200

	@@ -0,0 +1,3 @@

	@@ -0,0 +1,3 @@

	@@ -0,0 +1,3 @@

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:367a3576b66530fac1cee6c3f47b52e7bbfb003de6cc18e2e2c9a2cc2ad89fe6
+size 8383418560

+version https://git-lfs.github.com/spec/v1
+oid sha256:312aeab892aa958c06188775125a011164990e9c07acef75c6bd5208bd9bbaaf
+size 9267499200

+version https://git-lfs.github.com/spec/v1
+oid sha256:5652b9be0ea4ae2842130d04fe31bc869fcb99a2b7106c53b4e754a343fd688f
+size 9053114560

+version https://git-lfs.github.com/spec/v1
+oid sha256:b935eaeceeb165ac8c4786a6da44bcb713a8a253abccf01658bfb64acd4b8815
+size 10151579840

+version https://git-lfs.github.com/spec/v1
+oid sha256:1bc2d63bd2fa376f510850365e6826247d211270fbbbde08629a4c52c0c7aad4
+size 11035660480

+version https://git-lfs.github.com/spec/v1
+oid sha256:1cf76820e5b9b7ae11dafe82e0769e5564754616384c08fc3d1372eb06683aec
+size 10604187840

+version https://git-lfs.github.com/spec/v1
+oid sha256:a70147a8a8adb24d62a28e5038aca78c4c145a1c3043a38e8634738a9e6bbd93
+size 12030251200

+version https://git-lfs.github.com/spec/v1
+oid sha256:f68df5d3a22a01ea9ec547741f42ee7fedbf949bf42ef526fb1944fa0cfaa110
+size 15580500160

+version https://git-lfs.github.com/spec/v1
+oid sha256:ae08aeb1cb5da565300176b90c047d27d89dfe7d5f87a359ff95565c04a14102
+size 3591098560

+version https://git-lfs.github.com/spec/v1
+oid sha256:c3ed993354f92e1bd4bd1c64e879bb14c085811c817ca1aa9ffac4cc7f6fa665
+size 4230074560

+version https://git-lfs.github.com/spec/v1
+oid sha256:c0cdb8adae6864ae0b29cbd38dad2ea5952fee36321924d694a29e7f2eea0650
+size 29323399360