Instructions to use Kakyoin03/car-damage-assessment-llama-vision with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Kakyoin03/car-damage-assessment-llama-vision with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Kakyoin03/car-damage-assessment-llama-vision") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Kakyoin03/car-damage-assessment-llama-vision", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Kakyoin03/car-damage-assessment-llama-vision with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Kakyoin03/car-damage-assessment-llama-vision" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Kakyoin03/car-damage-assessment-llama-vision", "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/Kakyoin03/car-damage-assessment-llama-vision
- SGLang
How to use Kakyoin03/car-damage-assessment-llama-vision with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Kakyoin03/car-damage-assessment-llama-vision" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Kakyoin03/car-damage-assessment-llama-vision", "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 images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Kakyoin03/car-damage-assessment-llama-vision" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Kakyoin03/car-damage-assessment-llama-vision", "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" } } ] } ] }' - Unsloth Desktop
- Docker Model Runner
How to use Kakyoin03/car-damage-assessment-llama-vision with Docker Model Runner:
docker model run hf.co/Kakyoin03/car-damage-assessment-llama-vision
Car Damage Assessment Model
This model is a fine-tuned version of meta-llama/Llama-3.2-11B-Vision-Instruct specialized for professional car damage assessment.
Model Description
This AI model can analyze car images and provide detailed damage assessments including:
- Damage Detection: Identifies specific damaged parts (bumper, hood, doors, etc.)
- Severity Classification: Categorizes damage as minor, moderate, or major
- Professional Reports: Generates insurance-quality damage descriptions
- Multi-language Support: Works with English and French terminology
Training Details
- Base Model: Llama-3.2-11B-Vision-Instruct (11B parameters)
- Training Method: LoRA (Low-Rank Adaptation) fine-tuning
- Dataset: 5,600 car damage assessment samples
- Training Time: 138 minutes on NVIDIA L40S
- Final Loss: 0.1478
- Trainable Parameters: 67,174,400 out of 10,737,395,235 total
Usage
from unsloth import FastVisionModel
import torch
# Load model and tokenizer
model, tokenizer = FastVisionModel.from_pretrained("Kakyoin03/car-damage-assessment-llama-vision")
# Prepare image and prompt
from PIL import Image
image = Image.open("car_damage.jpg")
instruction = "You are a professional car damage assessment expert. Analyze this image and provide a detailed damage report."
messages = [
{"role": "user", "content": [
{"type": "image"},
{"type": "text", "text": instruction}
]}
]
# Generate assessment
input_text = tokenizer.apply_chat_template(messages, add_generation_prompt=True)
inputs = tokenizer(image, input_text, return_tensors="pt")
with torch.no_grad():
output = model.generate(**inputs, max_new_tokens=200)
assessment = tokenizer.decode(output[0], skip_special_tokens=True)
print(assessment)
Applications
- Insurance Claims: Automated damage assessment for insurance companies
- Auto Repair: Quick damage evaluation for repair shops
- Car Dealerships: Pre-purchase vehicle condition reports
- Mobile Apps: Real-time car damage assessment tools
Performance
The model demonstrates excellent performance in:
- Accurate damage part identification
- Reliable severity classification
- Detailed professional descriptions
- Multi-language damage terminology
- Fast inference (< 1 minute per assessment)
License
This model is released under the Llama 3.2 license. Please ensure compliance with the original license terms.
Acknowledgments
- Unsloth: For efficient training framework
- Meta: For the base Llama-3.2-Vision model
- Dataset: KHAOULA-KH/CAR_DOMMAGE_ for training data
Developed with
- Unsloth Framework
- PyTorch 2.4.0+cu121
- Transformers Library
- NVIDIA L40S GPU (44.7GB VRAM)
Trained on August 08, 2025 using advanced vision-language fine-tuning techniques.
Model tree for Kakyoin03/car-damage-assessment-llama-vision
Base model
meta-llama/Llama-3.2-11B-Vision-Instruct