MaliJa/Qwen2.5-Coder-7B-Instruct-openvino

Description

This model was converted to OpenVINO from Qwen/Qwen2.5-Coder-7B-Instruct using optimum-intel via the export space.

Quantization Parameters

Weight compression was performed in INT8_ASYM mode.

For more information on quantization, check the OpenVINO model optimization guide.

Running Model Inference with Optimum Intel

First make sure you have optimum-intel installed:

pip install optimum[openvino]

To run the model with optimum-intel:

from optimum.intel.openvino import OVModelForCausalLM
from transformers import AutoTokenizer

model_dir = "MaliJa/Qwen2.5-Coder-7B-Instruct-openvino"

tokenizer = AutoTokenizer.from_pretrained(model_dir)
model = OVModelForCausalLM.from_pretrained(
    model_dir,
    device="GPU",
)

prompt = "What is OpenVINO?"

inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=200)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))