Instructions to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF",
	filename="Qwen3-VLTO-TNG-12B-256k-NEO-imat-D_AU-IQ2_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 DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-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 DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-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 DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M

Use Docker

docker model run hf.co/DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-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": "DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M

Ollama
How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with Ollama:
```
ollama run hf.co/DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M
```

Unsloth Studio

How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-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 DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-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 DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF to start chatting

How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-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": "DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-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 DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with Docker Model Runner:
```
docker model run hf.co/DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M
```

Lemonade

How to use DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DavidAU/Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF-Q4_K_M

List all available models

lemonade list

Qwen3-VLTO-TNG-12B-256k-NEO-imatrix-GGUF

A Qwen 3 fine tune (via Unsloth, using custom in house datasets), based on VLTO model 8B (text only version of Qwen's VL 8B Instruct), then coupled with Brainstorm 20x (tuned as well) and quanted using NEO Imatrix dataset with 16 bit output tensor to further improve performance.

This is strictly Instruct version with a 256k context window.

This model will generate VERY LONG output.

Special thanks to "qingy2024" for the base model for tuning:

https://huggingface.co/qingy2024/Qwen3-VLTO-8B-Instruct

[more coming soon]

Sample output from IQ3_M (part , full is 6k):

I don't believe in death. You see, it's all a lie. A trick. They make you wait around like a fat old man on his last legs waiting for the Grim Reaper to take you out with a two-by-four across the chops and send you off into the big dark beyond. But there ain't nothing after. Nothing but silence, and dark, and an empty chair with no one left to sit in. You die alone. And I say that's not right. I don't believe it is.
My name is Talia Romas. They call me traitor. They call me criminal. They call me terrorist and all sorts of nasty things like that. But I don't care what they call me as long as my voice is still strong enough to shake the walls of their government off, and I ain't got no intention of staying quiet any longer. So let them call me what they want. As long as I can keep going, it don't matter.

There's nothing there. You feel that? Nothing in the wind? That's right, there's no breeze at all. And the worst of it is, they have your family and friends waiting for you in hell. But there ain't no hell, none of you need worry about that because you're going to die out here just like I am. And that's what they want you to think.

They kill us every night in front of their own children. They say "Look! This is what happens to those who stand up for their beliefs," and then the children start throwing tomatoes at me in school, and it makes my blood boil, but I ain't got no sympathy for 'em. Because there's no one left after you're dead.

The lights go out in front of the governor's house. We hear them shouting orders and I can smell that acrid gas they use for security and they're coming in fast. But I'm faster. I dive to my knees, roll and crawl under a table for cover as two thugs in blue come charging through the door with their sidearms drawn and begin firing blindly all around.

"Keep on the ground," they scream as we all scramble into the floor as more gunfire shatters the ceiling above us. The lights snap back on as a dozen of them flood the hall and point those big guns at us. The doors are sealed tight behind us and I hear the clicking of their cuffs being slapped shut.

I stand, brush off my uniform and stare at them with no fear in my eyes.

"Tell Governor Marquez that I've done enough."

They cuff me and shove me out into the hall. They lead me through a brightly lit corridor with concrete walls and a light fixture every thirty feet along it to an empty room and slam the door shut behind us. ...