Famous Vision Language Models and Their Architectures

Paddle Multimodal Integration and eXploration, supporting mainstream multi-modal tasks, including end-to-end large-scale multi-modal pretrain models and diffusion model toolbox. Equipped with high performance and flexibility.

A minimal codebase for finetuning large multimodal models, supporting llava-1.5/1.6, llava-interleave, llava-next-video, llava-onevision, qwen-vl, qwen2-vl, phi3-v etc.

【grps接入trtllm】通过接入TensorRT-LLM以及Tokenizers.cpp实现纯c++版本高性能LLM服务，兼容OpenAI接口协议，支持chat和function call模式，支持ai agent，支持分布式多卡推理，支持多模态，支持gradio聊天界面。

Mark web pages for use with vision-language models

Qwen-VL base model for use with Autodistill.