Instructions to use prashanth058/qwen2vl-flickr-lora-tower with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use prashanth058/qwen2vl-flickr-lora-tower with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-VL-2B-Instruct")
model = PeftModel.from_pretrained(base_model, "prashanth058/qwen2vl-flickr-lora-tower")

Transformers

How to use prashanth058/qwen2vl-flickr-lora-tower with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="prashanth058/qwen2vl-flickr-lora-tower")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("prashanth058/qwen2vl-flickr-lora-tower", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use prashanth058/qwen2vl-flickr-lora-tower with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prashanth058/qwen2vl-flickr-lora-tower"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prashanth058/qwen2vl-flickr-lora-tower",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/prashanth058/qwen2vl-flickr-lora-tower

SGLang

How to use prashanth058/qwen2vl-flickr-lora-tower 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 "prashanth058/qwen2vl-flickr-lora-tower" \
    --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": "prashanth058/qwen2vl-flickr-lora-tower",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "prashanth058/qwen2vl-flickr-lora-tower" \
        --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": "prashanth058/qwen2vl-flickr-lora-tower",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use prashanth058/qwen2vl-flickr-lora-tower with Docker Model Runner:
```
docker model run hf.co/prashanth058/qwen2vl-flickr-lora-tower
```

small doubt

by aastha6 - opened Nov 20, 2025

Discussion

aastha6

Nov 20, 2025

•

edited Nov 20, 2025

Is the difference between lora adapters for language vs tower vs tower connector is just the target modules? Is it possible to share the LoRA finetuning code? Also how do we evaluate the quality difference among the three when merged with base? It would be very helpful - I’m trying to replicate the same for Qwen2.5-VL and Qwen3-VL.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment