TIGER-Lab

/

VisCoder2-32B

like 1

Follow

TIGER-Lab 431

This PR improves the model card for VisCoder2-32B by adding two important metadata tags:

• library_name: transformers: The config.json file ("architectures": ["Qwen2ForCausalLM"], "model_type": "qwen2", "transformers_version": "4.43.1") provides clear evidence of compatibility with the Hugging Face Transformers library. This addition will enable an automated "How to use" code snippet on the model page for seamless integration.
• pipeline_tag: image-text-to-text: This tag accurately reflects the model's functionality as a multi-language visualization coding agent that processes multimodal inputs (inferred image context like UI screenshots alongside natural language instructions) to generate textual output (visualization code). This will enhance the model's discoverability under the relevant pipeline filter on the Hub.

A sample usage section is not included as the provided GitHub README does not contain a direct inference code snippet.