This Pull Request enhances the model card for `OpenGVLab/InternViT-300M-448px-V2_5` by:
- Adding a prominent link to the foundational paper for InternVL3.5: [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://huggingface.co/papers/2508.18265).
- Updating the `pipeline_tag` to `zero-shot-image-classification` to better reflect a key capability of the InternVL family, making it discoverable at https://huggingface.co/models?pipeline_tag=zero-shot-image-classification.
- Adding `library_name: transformers` to leverage the Hugging Face `transformers` library's automated usage snippets, improving accessibility for users.
- Including a direct link to the project page (https://chat.intern-ai.org.cn/) in the main content for easier access to related resources and demos.
- Updating the Citation section to include the `InternVL3.5` paper.
These changes aim to improve the model's discoverability, usability, and contextual information…
