# multimodal

BAGEL is a scalable unified multi-modal model that is revolutionizing the way AI interacts with complex systems. The model has dialogue reasoning, image generation, editing, style transfer, navigation, composition, thinking, and other functions, which provide a foundation for generating high-fidelity and realistic images by pretraining on large-scale alternating video and web data.

InternVL2-8B-MPO is a multimodal large language model (MLLM) that enhances multimodal inference capabilities by introducing a Mixed Preference Optimization (MPO) process. The model features an automated pipeline for preference data construction and builds the MMPR, a large-scale multimodal inference preference dataset. Based on the InternVL2-8B model, InternVL2-8B-MPO is fine-tuned using the MMPR dataset, demonstrating stronger multimodal inference capabilities with fewer hallucinations. The model achieved an accuracy of 67.0% on MathVista, surpassing the InternVL2-8B by 8.7 points, and performing closely to the much larger InternVL2-76B model.

InternVL 2.5 is an advanced multimodal large language model series based on InternVL 2.0. While maintaining the core model architecture, it introduces significant enhancements in training and testing strategies as well as data quality. This model explores the relationship between model scalability and performance, systematically investigating performance trends across visual encoders, language models, dataset sizes, and test settings. Comprehensive evaluations across a wide range of benchmarks, including interdisciplinary reasoning, document understanding, multi-image/video comprehension, real-world understanding, multimodal hallucination detection, visual localization, multilingual capabilities, and pure language processing, demonstrate InternVL 2.5's competitiveness comparable to leading commercial models like GPT-4o and Claude-3.5-Sonnet. Notably, it is the first open-source MLLM to achieve over 70% on the MMMU benchmark, attaining a 3.7 percentage point improvement through Chain of Thought (CoT) reasoning, showcasing strong potential for scalability during testing.

Gemini Pro is a high-performance multimodal AI model launched by DeepMind, designed for a wide range of tasks. It features a long context window of up to two million tokens and can handle large-scale documents, code, audio, and video. It excels in various benchmark tests, including code generation, solving mathematical problems, and multilingual translation.

Vidu, co-released by Shengshu Technology and Tsinghua University, is the first long-duration, high-consistency, and high-dynamic video large model in China. This model utilizes a proprietary architecture, U-ViT, which merges Diffusion with Transformer, supporting one-click generation of up to 16-second videos with 1080P resolution. Vidu not only simulates the real physical world but also boasts rich imagination, characteristics such as multi-camera generation, and high temporal-spatial consistency. Its rapid breakthrough is attributed to the team's long-term accumulation in Bayesian machine learning and multimodal large models, as well as numerous original achievements. Vidu's launch represents the sustained innovative capabilities and leadership of Shengshu Technology in the multimodal native large model field. Looking to the future, its flexible architecture will be able to accommodate a wider range of modalities, further expanding the boundaries of multimodal general capabilities.