Instructions to use alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf",
	filename="model-q4_k_m.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": [
				{
					"type": "text",
					"text": "Describe this image in one sentence."
				},
				{
					"type": "image_url",
					"image_url": {
						"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
					}
				}
			]
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M

Use Docker

docker model run hf.co/alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "alvarobrito/lamparinalm-qwen3-vl-pt-br-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": "alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M

Ollama
How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf with Ollama:
```
ollama run hf.co/alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M
```

Unsloth Studio

How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf to start chatting

How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf alvarobrito/lamparinalm-qwen3-vl-pt-br-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": "alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-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 alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf with Docker Model Runner:
```
docker model run hf.co/alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M
```

Lemonade

How to use alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf:Q4_K_M

Run and chat with the model

lemonade run user.lamparinalm-qwen3-vl-pt-br-gguf-Q4_K_M

List all available models

lemonade list

LamparinaLLM — Qwen3-VL 2B Fine-tuned PT-BR

Uma lamparina não ilumina o mundo — mas ilumina o que você precisa.

LamparinaLLM é um modelo de visão-linguagem fine-tunado em português brasileiro, parte da série de ebooks "Criando seu próprio SLM do Zero" de Álvaro Brito.

Este repositório contém o GGUF Q4_K_M para deploy local com ollama. Os adaptadores LoRA estão em alvarobrito/lamparinalm-qwen3-vl-pt-br.

Como usar

# Baixar e rodar com ollama
ollama run hf.co/alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf

# Enviar uma imagem
ollama run hf.co/alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf "Descreva esta imagem" --image foto.jpg

import ollama

resposta = ollama.chat(
    model='hf.co/alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf',
    messages=[{
        'role': 'user',
        'content': 'O que você vê nesta imagem?',
        'images': ['./foto.jpg'],
    }],
)
print(resposta['message']['content'])

Fine-tuning

Parâmetro	Valor
Modelo base	`Qwen/Qwen3-VL-2B-Instruct`
Método	QLoRA (4-bit NF4) + RSLoRA
LoRA r / alpha	16 / 32
Steps	500
Batch efetivo	8
Learning rate	0.0002
Biblioteca	Unsloth
Plataforma	Kaggle T4 (gratuito)

Datasets de Treinamento

Dataset	Tipo	Uso
PraCegoVer Filtrado FSB	Imagem + legenda PT-BR	Alinhamento visual em português
FM30K	Imagem + legenda PT-BR	Legendas estruturadas (FrameNet Brasil)
Alpaca PT-BR	Instrução texto	Seguimento de instrução em PT-BR

Sobre a Série

Este modelo foi criado como parte do livro "SLM do Zero — v3: O Inimigo Agora é Outro" de Álvaro Brito. A série ensina a construir, aprofundar e adaptar modelos de linguagem usando apenas recursos gratuitos (Kaggle T4, Hugging Face).

Track A: modelo multimodal 52M parâmetros construído do zero com SigLIP + projetor MLP
Track B (este modelo): fine-tuning QLoRA de VLM de produção em PT-BR

Licença

Apache 2.0 — herdada do modelo base Qwen3-VL 2B.

Downloads last month: 406

GGUF

Model size

2B params

Architecture

qwen3vl

Hardware compatibility

4-bit

Model tree for alvarobrito/lamparinalm-qwen3-vl-pt-br-gguf

Base model

Qwen/Qwen3-VL-2B-Instruct

Quantized

(74)

this model

alvarobrito
/

lamparinalm-qwen3-vl-pt-br-gguf