Instructions to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="SandLogicTechnologies/Phi-4-mini-reasoning-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("SandLogicTechnologies/Phi-4-mini-reasoning-GGUF", device_map="auto")

llama-cpp-python

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="SandLogicTechnologies/Phi-4-mini-reasoning-GGUF",
	filename="Phi-4-mini-reasoning-Q4_K_M.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF: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 SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF: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 SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

Use Docker

docker model run hf.co/SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SandLogicTechnologies/Phi-4-mini-reasoning-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": "SandLogicTechnologies/Phi-4-mini-reasoning-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

SGLang

How to use SandLogicTechnologies/Phi-4-mini-reasoning-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 "SandLogicTechnologies/Phi-4-mini-reasoning-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": "SandLogicTechnologies/Phi-4-mini-reasoning-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 "SandLogicTechnologies/Phi-4-mini-reasoning-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": "SandLogicTechnologies/Phi-4-mini-reasoning-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with Ollama:
```
ollama run hf.co/SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M
```

Unsloth Studio

How to use SandLogicTechnologies/Phi-4-mini-reasoning-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 SandLogicTechnologies/Phi-4-mini-reasoning-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 SandLogicTechnologies/Phi-4-mini-reasoning-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for SandLogicTechnologies/Phi-4-mini-reasoning-GGUF to start chatting

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with Docker Model Runner:
```
docker model run hf.co/SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M
```

Lemonade

How to use SandLogicTechnologies/Phi-4-mini-reasoning-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SandLogicTechnologies/Phi-4-mini-reasoning-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Phi-4-mini-reasoning-GGUF-Q4_K_M

List all available models

lemonade list

Phi-4-Mini-Reasoning (GGUF Q4_KM) - Sandlogic Lexicons

Model Summary

Phi-4-Mini-Reasoning is a lightweight open-source model from the Phi-4 family, designed with a strong focus on high-quality, reasoning-dense synthetic data. It has been further fine-tuned for advanced mathematical reasoning tasks and supports a 128K token context length. This model is especially optimized for logic-intensive scenarios while maintaining a compact size, making it ideal for memory and compute-constrained environments.

Model Family: Phi-4
Parameter Count: 3.8B
Architecture: Dense decoder-only Transformer
Context Length: 128K tokens
Quantization: GGUF Q4_KM
Supported Language: English
Release Date: April 2025
Cutoff Date: February 2025

Intended Uses

Primary Use Cases

Phi-4-Mini-Reasoning is designed to excel at:

Multi-step mathematical reasoning
Formal proof generation
Symbolic computation
Solving advanced word problems
Tasks requiring structured logic and analytical thinking

Its high context length and reasoning capabilities make it suitable for latency-bound applications and deployments on resource-constrained hardware.

Use Case Considerations

This model is optimized specifically for mathematical reasoning tasks.
It is not evaluated for general-purpose downstream tasks such as conversational AI or creative writing.
Developers should:
- Assess use case suitability.
- Account for limitations in multi-language support.
- Evaluate performance, safety, and fairness—especially in high-risk or regulated environments.
- Ensure compliance with all applicable laws and regulations (e.g., privacy and trade compliance).

Training Details

Model Architecture: Same as Phi-4-Mini with 3.8B parameters
Notable Enhancements:
- 200K vocabulary
- Grouped-query attention
- Shared input/output embeddings
Training Dataset Size: 150B tokens
Training Duration: 2 days
Hardware Used: 128 × H100-80G GPUs
Training Date: February 2024
Output: Generated text
Input Format: Text (chat-style prompts recommended)

Integration in Lexicons

This quantized GGUF Q4_KM version of Phi-4-Mini-Reasoning is included in our Sandlogic Lexicons model zoo, making it readily available for efficient inference in edge deployments and research use cases focused on math reasoning.

For optimal results, we recommend using Phi-4-Mini-Reasoning in tasks that require deep mathematical analysis and structured problem solving.

Downloads last month: 17

GGUF

Model size

4B params

Architecture

phi3

Hardware compatibility

4-bit

Model tree for SandLogicTechnologies/Phi-4-mini-reasoning-GGUF

Base model

microsoft/Phi-4-mini-reasoning

Quantized

(35)

this model