Ir para o conteúdo
ModelListadosExterno
Baixar

Autopilot Qwen3 Vl

Autopilot-Qwen3-VL is an end-to-end autonomous driving model built on top of the powerful Qwen3-VL-2B-Instruct Vision-Language Model. It takes a single road/dashcam image as input and directly predicts the vehicle's...

Fontehuggingface.co/Aleton/Autowheel-qwen3-vl/tree/e1d4a806dc3d0c8aab5ac3e27ee16db5d02f5abb
TipoModeloModelo baseQwen/Qwen3-VL-2B-InstructVersãoRevision e1d4a80Bnota

Fonte do modelo

Tipo
Modelo
Tamanho do arquivo
4,02 GB
Modelo base
Qwen/Qwen3-VL-2B-Instruct
Versão
vRevision e1d4a80
Atualizado
30 de out. de 2026
Downloads
20
Fonte
Hugging Face

Descrição da fonte

🚗 Autopilot-Qwen3-VL

Autopilot-Qwen3-VL is an end-to-end autonomous driving model built on top of the powerful Qwen3-VL-2B-Instruct Vision-Language Model. It takes a single road/dashcam image as input and directly predicts the vehicle's continuous control parameters: target speed (km/h) and steering torque (N).

🎥 Simulation Demo

Simulation Demo

🧠 Model Details

The model utilizes a custom regression head on top of the frozen Qwen3-VL base, trained using Parameter-Efficient Fine-Tuning (PEFT/LoRA) for optimal performance and resource efficiency.

  • Base Model: Qwen/Qwen3-VL-2B-Instruct
  • Total Parameters: ~2.13B (2,132,320,770)
  • Trainable Parameters: ~4.78M (4,788,738 / 0.225%)
  • Architecture Type: Vision-Language Model + Dual Regression Head

Architecture Diagram

📊 Dataset & Output Format

Trained on the SADC Situation Awareness Dataset.

⚠️ Important Note on Steering Values

Based on the dataset's coordinate system (standard automotive physics):

  • Negative values (-) = Steering RIGHT
  • Positive values (+) = Steering LEFT

🚀 Usage

To run inference, you need the custom autopilot_inference.py script provided in this repository.

1. Download the inference script

from huggingface_hub import hf_hub_download

hf_hub_download(
    repo_id="Aleton/Autopilot-qwen3-vl", 
    filename="autopilot_inference.py", 
    local_dir="."
)

2. Run Inference

from autopilot_inference import AutopilotInference
from PIL import Image

# 1. Load the model (downloads weights automatically)
autopilot = AutopilotInference.from_pretrained("Aleton/Autopilot-qwen3-vl")

# 2. Load a dashcam image
image = Image.open("road.jpg")

# 3. Get predictions
result = autopilot.predict(image)

print(f"Target Speed: {result['speed_kmh']:.1f} km/h")
print(f"Steering Torque: {result['steering_N']:.3f} N")

⚠️ Disclaimer

This model is built for educational and research purposes only. It is not designed, tested, or certified for use in real-world autonomous vehicles. Never rely on this model to control a real car.

Fontes

1 fonte
Hugging FacePrimárioVersão e1d4a806dc3d0c8aab5ac3e27ee16db5d02f5abb
huggingface.co/Aleton/Autowheel-qwen3-vl/tree/e1d4a806dc3d0c8aab5ac3e27ee16db5d02f5abb

Verificado 30 de out.

Artefatos de modelo

1 artefato

model.pt

pt · 4,02 GB · SHA-256 976ec4de8aa8…7cba · Hugging Face

Download file

Trechos de fonte

3 trechos
Readme14 de ago.huggingface.co/Aleton/Autopilot-qwen3-vl

--- license: mit language: - en tags: - autonomous-driving - vision-language-model - computer-vision - regression base_model: Qwen/Qwen3-VL-2B-Instruct datasets: - >- jHaselberger/SADC-Situation-Awareness-for-Driver-Centric-Driving-Style-Adaptation pipeline_tag: image-to-text --- # 🚗 Autopilot-Qwen3-VL **Autopilot-Qwen3-VL** is an end-to-end autonomous driving model built on top of the powerful `Qwen3-VL-2B-Instruct` Vision-Language Model. It takes a single road/dashcam image as input and directly predicts the vehicle's continuous control parameters: **target speed (km/h)** and **steering torque (N)**. ## 🎥 Simulation Demo ![Simulation Demo](https://huggingface.co/Aleton/Autopilot-qwen3-vl/resolve/main/demo.gif) --- ## 🧠 Model Details The model utilizes a custom regression head on top of the frozen Qwen3-VL base, trained using Parameter-Efficient Fine-Tuning (PEFT/LoRA) for optimal performance and resource efficiency. - **Base Model:** `Qwen/Qwen3-VL-2B-Instruct` - **Total Parameters:** ~2.13B (2,132,320,770) - **Trainable Parameters:** ~4.78M (4,788,738 / 0.225%) - **Architecture Type:** Vision-Language Model + Dual Regression Head ![Architecture Diagram](architecture.jpg) ## 📊 Dataset & Output Format Trained on the [SADC Situation Awareness Dataset](https://huggingface.co/datasets/jHaselberger/SADC-Situation-Awareness-for-Driver-Centric-Driving-Style-Adaptation). ### ⚠️ Important Note on Steering Values Based on the dataset's coordinate system (standard automotive physics): - **Negative values (`-`)** = Steering **RIGHT** - **Positive values (`+`)** = Steering **LEFT** --- ## 🚀 Usage To run inference, you need the custom `autopilot_inference.py` script provided in this repository. ### 1. Download the inference script ```python from huggingface_hub import hf_hub_download hf_hub_download( repo_id="Aleton/Autopilot-qwen3-vl", filename="autopilot_inference.py", local_dir="." ) ``` ### 2. Run Inference ```python from autopilot_inference import AutopilotInference from PIL import Image # 1. Load the model (downloads weights automatically) autopilot = AutopilotInference.from_pretrained("Aleton/Autopilot-qwen3-vl") # 2. Load a dashcam image image = Image.open("road.jpg") # 3. Get predictions result = autopilot.predict(image) print(f"Target Speed: {result['speed_kmh']:.1f} km/h") print(f"Steering Torque: {result['steering_N']:.3f} N...

Note30 de out.huggingface.co/Aleton/Autowheel-qwen3-vl

Source context: 20 downloads · 1 likes · Pipeline image-to-text · Repo Aleton/Autowheel-qwen3-vl

Note14 de ago.huggingface.co/Aleton/Autopilot-qwen3-vl

Source context: 16 downloads · 0 likes · Pipeline image-to-text · Repo Aleton/Autopilot-qwen3-vl