Optimized Model: Qwen3-VL-2B-Instruct-4bit-bnb

This is an optimized version of Qwen/Qwen3-VL-2B-Instruct.

Optimizations Applied

• Quantize to 4-bit (BNB)

How to Use

from transformers import AutoProcessor, AutoModelForCausalLM
model_id = "broadfield-dev/Qwen3-VL-2B-Instruct-4bit-bnb"
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", trust_remote_code=True)

ONNX Information

This model was converted to ONNX with the task: vision2seq-lm.