Instructions to use KoarAI/LFM2.5-350M-Thinking-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 KoarAI/LFM2.5-350M-Thinking-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 KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf KoarAI/LFM2.5-350M-Thinking-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 KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf KoarAI/LFM2.5-350M-Thinking-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 KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf KoarAI/LFM2.5-350M-Thinking-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 KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M

Use Docker

docker model run hf.co/KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use KoarAI/LFM2.5-350M-Thinking-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KoarAI/LFM2.5-350M-Thinking-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": "KoarAI/LFM2.5-350M-Thinking-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M

Ollama
How to use KoarAI/LFM2.5-350M-Thinking-GGUF with Ollama:
```
ollama run hf.co/KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M
```
Unsloth Desktop

How to use KoarAI/LFM2.5-350M-Thinking-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf KoarAI/LFM2.5-350M-Thinking-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": "KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use KoarAI/LFM2.5-350M-Thinking-GGUF with Docker Model Runner:
```
docker model run hf.co/KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M
```

Lemonade

How to use KoarAI/LFM2.5-350M-Thinking-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.LFM2.5-350M-Thinking-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use KoarAI/LFM2.5-350M-Thinking-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 KoarAI/LFM2.5-350M-Thinking-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 KoarAI/LFM2.5-350M-Thinking-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use KoarAI/LFM2.5-350M-Thinking-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf KoarAI/LFM2.5-350M-Thinking-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 "KoarAI/LFM2.5-350M-Thinking-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"

🐨 KoarAI / LFM2.5-350M-Thinking-GGUF

🌟 Overview

This repository contains official GGUF format quantizations of KoarAI/LFM2.5-350M-Thinking, an ultra-lightweight reasoning model trained with Full Fine-Tuning on multi-teacher reasoning distillation traces (Qwen 3.8 Max, GLM 5.2, and Kimi K3).

These GGUF files are ready for high-speed inference across llama.cpp, Ollama, LM Studio, Jan.ai, and mobile/edge devices.

📦 Available Quantizations

File	Format	Quantization Detail	Size	Recommended Use Case
`LFM2.5-350M-Thinking-f16.gguf`	`F16`	Full unquantized 16-bit float	~710 MB	Maximum accuracy, baseline
`LFM2.5-350M-Thinking-Q8_0.gguf`	`Q8_0`	High quality 8-bit quantization	~375 MB	Near-lossless reasoning fidelity
`LFM2.5-350M-Thinking-Q5_K_M.gguf`	`Q5_K_M`	Balanced 5-bit k-quant	~260 MB	Great balance of speed & reasoning
`LFM2.5-350M-Thinking-Q4_K_M.gguf`	`Q4_K_M`	Recommended 4-bit k-quant	~220 MB	Best default for laptops and edge
`LFM2.5-350M-Thinking-Q4_0.gguf`	`Q4_0`	Fast standard 4-bit	~205 MB	Ultra-fast execution on CPU

⚡ Quickstart with Ollama

Create a Modelfile:

FROM ./LFM2.5-350M-Thinking-Q4_K_M.gguf

TEMPLATE """<|im_start|>system
{{ .System }}<|im_end|>
{{- range .Messages }}
<|im_start|>{{ .Role }}
{{ .Content }}<|im_end|>
{{- end }}
<|im_start|>assistant
<think>
"""

PARAMETER stop "<|im_end|>"
PARAMETER temperature 0.6
PARAMETER top_p 0.9

Run in terminal:

ollama create koarai-350m -f Modelfile
ollama run koarai-350m "How many 'r' in strawberry?"

💻 Quickstart with llama.cpp

./llama-cli -m ./LFM2.5-350M-Thinking-Q4_K_M.gguf \
  -p "<|im_start|>user\nSolve: 2x + 10 = 24<|im_end|>\n<|im_start|>assistant\n<think>\n" \
  -n 512 --temp 0.6