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

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use liodon-ai/Phi-4-mini-instruct-imatrix-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 liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf liodon-ai/Phi-4-mini-instruct-imatrix-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 liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf liodon-ai/Phi-4-mini-instruct-imatrix-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 liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf liodon-ai/Phi-4-mini-instruct-imatrix-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 liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M

Use Docker

docker model run hf.co/liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use liodon-ai/Phi-4-mini-instruct-imatrix-GGUF with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M

Ollama
How to use liodon-ai/Phi-4-mini-instruct-imatrix-GGUF with Ollama:
```
ollama run hf.co/liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
```
Unsloth Desktop

How to use liodon-ai/Phi-4-mini-instruct-imatrix-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M

Configure the model in Pi

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

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use liodon-ai/Phi-4-mini-instruct-imatrix-GGUF with Docker Model Runner:
```
docker model run hf.co/liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
```

Lemonade

How to use liodon-ai/Phi-4-mini-instruct-imatrix-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Phi-4-mini-instruct-imatrix-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use liodon-ai/Phi-4-mini-instruct-imatrix-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 liodon-ai/Phi-4-mini-instruct-imatrix-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 liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use liodon-ai/Phi-4-mini-instruct-imatrix-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf liodon-ai/Phi-4-mini-instruct-imatrix-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 "liodon-ai/Phi-4-mini-instruct-imatrix-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"

PY-AI-Dev commited on Jul 9

Commit

d0240d4

verified ·

1 Parent(s): 2c942b0

Add iMatrix GGUF quantizations for Phi-4-mini-instruct

Browse files

Files changed (9) hide show

.gitattributes +7 -0
Phi-4-mini-instruct-IQ2_M.gguf +3 -0
Phi-4-mini-instruct-IQ3_M.gguf +3 -0
Phi-4-mini-instruct-IQ4_XS.gguf +3 -0
Phi-4-mini-instruct-Q4_K_M.gguf +3 -0
Phi-4-mini-instruct-Q5_K_M.gguf +3 -0
Phi-4-mini-instruct-Q6_K.gguf +3 -0
Phi-4-mini-instruct-Q8_0.gguf +3 -0
README.md +64 -0

.gitattributes CHANGED Viewed

@@ -33,3 +33,10 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+Phi-4-mini-instruct-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
+Phi-4-mini-instruct-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
+Phi-4-mini-instruct-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
+Phi-4-mini-instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+Phi-4-mini-instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+Phi-4-mini-instruct-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
+Phi-4-mini-instruct-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text

Phi-4-mini-instruct-IQ2_M.gguf ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:fa7bf8020ad7972eaeafd03f133b89762db7f4343a559500533a5472a4fe15a4
+size 1562475136

Phi-4-mini-instruct-IQ3_M.gguf ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:8b20c7edfaebb97509157675a9699da9d017233fd805f808eae9814e3f647749
+size 2017656448

Phi-4-mini-instruct-IQ4_XS.gguf ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:b7c2f00af33115a5e148e8da2d320cd0c29d06b02a9b91e95e5a889b1dda6e11
+size 2224488064

Phi-4-mini-instruct-Q4_K_M.gguf ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:348a036e5145465545eacc7aee6a383fca50b76e68cf90d10caa2ca51a5af592
+size 2493841024

Phi-4-mini-instruct-Q5_K_M.gguf ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:0df4deb00ebdbf46ac9b9f6a1e761f20b37dac4cd6ea3421281b17600ff53b0b
+size 2814705280

Phi-4-mini-instruct-Q6_K.gguf ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:f248b8065fc038fc08368ee73ff3272be6decf49605f9cd4139cacc5e8a105b8
+size 3155623552

Phi-4-mini-instruct-Q8_0.gguf ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:c63915f458d3f0f4b5bf73e110c744c62447f7d06284cd1864841bdfa91ce9c5
+size 4084611712

README.md ADDED Viewed

	@@ -0,0 +1,64 @@

+---
+license: other
+base_model: microsoft/Phi-4-mini-instruct
+pipeline_tag: text-generation
+tags:
+- gguf
+- local-llm
+- llama.cpp
+- lm-studio
+- quantized
+- imatrix
+- sub-4-bit
+- phi3
+---
+# Phi-4-mini-instruct — iMatrix GGUF
+GGUF quantizations of [microsoft/Phi-4-mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct), published by [Liodon AI](https://huggingface.co/liodon-ai).
+## Quick Start
+**llama.cpp**
+```bash
+llama-cli -hf liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
+```
+**Ollama**
+```bash
+ollama run hf.co/liodon-ai/Phi-4-mini-instruct-imatrix-GGUF:Q4_K_M
+```
+**LM Studio / Jan** — search `liodon-ai/Phi-4-mini-instruct-imatrix-GGUF` and pick your quant.
+## Quants
+| Quant | Size | VRAM est. | Notes |
+|-------|------|-----------|-------|
+| `IQ2_M` | 1.56 GB | ~2 GB | 2-bit, iMatrix — smallest usable |
+| `IQ3_M` | 2.02 GB | ~2 GB | 3-bit, iMatrix — great quality/size tradeoff |
+| `IQ4_XS` | 2.22 GB | ~3 GB | 4-bit extra-small, iMatrix |
+| `Q4_K_M` | 2.49 GB | ~3 GB | 4-bit, iMatrix-calibrated (recommended) |
+| `Q5_K_M` | 2.81 GB | ~3 GB | 5-bit, iMatrix-calibrated |
+| `Q6_K` | 3.16 GB | ~4 GB | 6-bit, iMatrix-calibrated, near-lossless |
+| `Q8_0` | 4.08 GB | ~5 GB | 8-bit, essentially lossless |
+## What is iMatrix?
+Standard quantization treats all weights equally. iMatrix runs 128 calibration chunks through
+the full-precision model to find which weights matter most, then allocates more precision where
+it counts. At Q2/Q3/Q4 this means noticeably better coherence and instruction-following —
+**same file size, better output**.
+Calibration: 2M tokens of [WikiText-103](https://huggingface.co/datasets/wikitext).
+> Also see plain (non-iMatrix) quants: `liodon-ai/Phi-4-mini-instruct-GGUF`
+## Source
+- **Model**: [microsoft/Phi-4-mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct)
+- **License**: other
+---
+*Quantized by [Liodon AI](https://huggingface.co/liodon-ai)*