[OpenVINO] Support Gemma3n

[rkazants]

What does this PR do?

The command for model exporting:

optimum-cli export openvino -m google/gemma-3n-E2B-it gemma-3n-E2B-it --task=image-text-to-text

The script for inference:

from transformers import AutoProcessor
from optimum.intel.openvino import OVModelForVisualCausalLM

model_id = "google/gemma-3n-E2B-it"
model = OVModelForVisualCausalLM.from_pretrained(model_id)

processor = AutoProcessor.from_pretrained(model_id, padding_side="left")

url = "https://media.istockphoto.com/id/1192867753/photo/cow-in-berchida-beach-siniscola.jpg?s=612x612&w=0&k=20&c=v0hjjniwsMNfJSuKWZuIn8pssmD5h5bSN1peBd1CmH4="
messages = [
    {
        "role": "system",
        "content": [
            {"type": "text", "text": "You are a helpful assistant."}
        ]
    },
    {
        "role": "user", "content": [
            {"type": "image", "url": url},
            {"type": "text", "text": "What is shown in this image?"},
        ]
    },
]
inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    return_dict=True,
    return_tensors="pt",
    add_generation_prompt=True,
)

output = model.generate(**inputs, max_new_tokens=50)
print(processor.decode(output[0, inputs.input_ids.shape[1]: ], skip_special_tokens=True))

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[Copilot]

Pull request overview

This PR adds support for the Gemma3n vision-language model to the OpenVINO integration in optimum-intel. Gemma3n appears to be a variant of Gemma3 with special attention mechanisms including sliding attention and shared KV caches.

Changes:

• Registers Gemma3n model type and configuration classes for OpenVINO export
• Adds custom KV cache handling for Gemma3n's unique architecture (sliding attention, shared layers)
• Maps Gemma3n to the same visual-language model class as Gemma3

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
optimum/intel/openvino/modeling_visual_language.py	Maps "gemma3n" to _OVGemma3ForCausalLM class for inference
optimum/exporters/openvino/utils.py	Adds "gemma3n" to the list of supported vision-language models
optimum/exporters/openvino/model_configs.py	Defines Gemma3nTextOpenVINOConfig with custom KV cache handling and Gemma3nOpenVINOConfig for vision-language export

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.