Request for LLaVA-OneVision Model & Data Specifications

[jungsunIm]

Hello,

We'd like to begin by expressing our sincere appreciation for your team's excellent work on LLaVA-OneVision and for making this powerful model publicly available. It is a fantastic contribution to the research community.

To ensure we are documenting and integrating the model correctly, we are writing to request official information regarding the following items. We have already reviewed the technical reports and the Hugging Face model card, but we would like to receive an official response for confirmation.

Could you please provide answers for the following?

(a) General description of the AI model

• Tasks the model is intended to perform and the types and characteristics of AI systems it can be integrated into
• Applicable use policies
• Release date and distribution methods
• Architecture and number of parameters
• Input and output modalities and formats
• License for the model

(b) Description of processes and components for developing AI systems based on the AI model

• Technical means required to integrate the general-purpose AI model into an AI system (infrastructure, usage guidelines)
• Maximum input and output context size
• Information on the data used for training, testing, and validation, including data types, sources, and curation methodology

Thank you for your assistance. We look forward to your response.