Instructions to use EryriLabs/LFM2.5-VL-3B-DragOn-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 EryriLabs/LFM2.5-VL-3B-DragOn-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 EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf EryriLabs/LFM2.5-VL-3B-DragOn-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 EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf EryriLabs/LFM2.5-VL-3B-DragOn-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 EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf EryriLabs/LFM2.5-VL-3B-DragOn-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 EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M

Use Docker

docker model run hf.co/EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use EryriLabs/LFM2.5-VL-3B-DragOn-GGUF with Ollama:
```
ollama run hf.co/EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M
```
Unsloth Desktop

How to use EryriLabs/LFM2.5-VL-3B-DragOn-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf EryriLabs/LFM2.5-VL-3B-DragOn-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": "EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use EryriLabs/LFM2.5-VL-3B-DragOn-GGUF with Docker Model Runner:
```
docker model run hf.co/EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M
```

Lemonade

How to use EryriLabs/LFM2.5-VL-3B-DragOn-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.LFM2.5-VL-3B-DragOn-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use EryriLabs/LFM2.5-VL-3B-DragOn-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 EryriLabs/LFM2.5-VL-3B-DragOn-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 EryriLabs/LFM2.5-VL-3B-DragOn-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use EryriLabs/LFM2.5-VL-3B-DragOn-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf EryriLabs/LFM2.5-VL-3B-DragOn-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 "EryriLabs/LFM2.5-VL-3B-DragOn-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"

LFM2.5-VL-3B-DragOn — GGUF

GGUF quants of EryriLabs/LFM2.5-VL-3B-DragOn: LiquidAI's LFM2.5-VL-3B fine-tuned for drag-and-drop grounding on GUI screenshots. Screenshot + instruction in, {"start":[x,y],"end":[x,y]} out (0-1000 normalised coordinates).

The short version of the story: the base model scores 0.7% on the DragOn public eval, this fine-tune scores 70.8% acc@5 (78.5% acc@10), and it cost about $36 to train. Details, per-domain numbers and caveats are in the main repo.

Files

You need TWO files: a main model quant plus the vision projector (mmproj).

file	size	note
`LFM2.5-VL-3B-DragOn-Q4_K_M.gguf`	~1.5 GB	good default, runs on almost anything
`LFM2.5-VL-3B-DragOn-Q5_K_M.gguf`	~1.8 GB
`LFM2.5-VL-3B-DragOn-Q6_K.gguf`	~2.0 GB	recommended if you have the room
`LFM2.5-VL-3B-DragOn-Q8_0.gguf`	~2.7 GB
`LFM2.5-VL-3B-DragOn-F16.gguf`	~5.1 GB	reference
`mmproj-LFM2.5-VL-3B-DragOn-F16.gguf`	~0.8 GB	vision projector, always required

Note that coordinates are a precision task, so if you see degraded accuracy at Q4, step up a quant before blaming the model.

Usage

llama-server -m LFM2.5-VL-3B-DragOn-Q6_K.gguf \
  --mmproj mmproj-LFM2.5-VL-3B-DragOn-F16.gguf \
  -ngl 99 -c 4096 --temp 0

Then send a chat completion with the image and this exact prompt shape (it's what the model was trained on):

This is a screenshot of a user interface. You must perform a DRAG action.
Task: <your instruction>
Give the drag as JSON with the START point (where the mouse button goes down) and the END point (where it is released), in coordinates normalised to 0-1000 for both x (left->right) and y (top->bottom):
{"start":[x,y],"end":[x,y]}
Output only the JSON.

Multiply by your actual screen size /1000 and you have your drag.

Thanks

To LiquidAI for the base model and to Nathan Bout, Maxime Langevin and Ronan Riochet at Hcompany for the DragOn dataset — see the main repo card for the full credits.

Quantised by Dwain Barnes (EryriLabs), August 2026.

Downloads last month: 279

GGUF

Model size

3B params

Architecture

lfm2

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for EryriLabs/LFM2.5-VL-3B-DragOn-GGUF

Base model

LiquidAI/LFM2.5-2.6B-Base

Finetuned

LiquidAI/LFM2.5-VL-3B

Finetuned

EryriLabs/LFM2.5-VL-3B-DragOn

Quantized

(1)

this model

EryriLabs
/

LFM2.5-VL-3B-DragOn-GGUF

LFM2.5-VL-3B-DragOn — GGUF

Files

Usage

Thanks

Model tree for EryriLabs/LFM2.5-VL-3B-DragOn-GGUF

Dataset used to train EryriLabs/LFM2.5-VL-3B-DragOn-GGUF