# Multimodal interaction

The Llama 4 series is a native multimodal AI model launched by Meta. It adopts the Mixture of Experts architecture, supports text and image interaction, and performs excellently in various language and visual tasks.

The Llama 4 series is a native multimodal AI model launched by Meta, supporting text and image interaction. It adopts the Mixture of Experts architecture and performs excellently in text and image understanding.

Qwen2-VL-2B is a multimodal model that can handle image and text inputs and generate text outputs.

VideoChat-Online is an online video understanding model based on Phi-3-vision-128k-instruct, focusing on the video text-to-text task.

UGround is a powerful GUI visual positioning model trained with a simple recipe, developed in collaboration by OSU NLP Group and Orby AI.

PAE-LLaVa-7B is a foundation model Internet intelligent agent based on the PAE (Proposer-Agent-Evaluator) framework, focusing on autonomous skill discovery.

An Any-to-Any subnet model developed in collaboration by OMEGA Labs and Bittensor, supporting multiple task conversions

Mini-Omni2 is a fully interactive multimodal model capable of understanding image, audio, and text inputs, and engaging in end-to-end voice conversations with users.

A multimodal model that combines visual and language abilities, achieving image-text interaction through a merging method