AI Briefing#multimodal-ai
The latest AI and developer news about #multimodal-ai, with the original source and a short summary.
The latest AI and developer news about #multimodal-ai, with the original source and a short summary.
Google has enhanced multi-object visual search, enabling simultaneous search of multiple objects within an image using Gemini models.
Gemini 3.1 Flash Image Preview (Nano Banana 2) has launched on AI Gateway, offering enhanced visual quality and reasoning capabilities.
Netflix strengthened shot-level content understanding with MediaFM, which combines audio, video, and text.
Emotional expressiveness improved through DPO training and actor recording data, surpassing GPT-4o on Korean image-based benchmarks
ARHIS Ambient recorded a 96.7% detection accuracy rate (TDR) for black ice detection using multimodal AI that combines audio, video, and environmental data.
SafetyKit has significantly enhanced the accuracy and processing capacity of its multimodal risk agents by leveraging OpenAI's cutting-edge models, including GPT-5.
TimeScope, an open-source benchmark for measuring temporal understanding of long videos, has been released.
LG AI Research has released 'EXAONE Path 2.0', a next-generation cancer diagnostic AI model that predicts gene mutations from tissue images, as open source.
Qwen VLo is a multimodal model that combines image understanding and generation into one.
The visual reasoning model **QVQ-Max**, which analyzes visual information and reasons to present solutions, has been released.
Qwen2.5-Omni, the new flagship model in the Qwen series, is an end-to-end multimodal model that unifiedly processes text, images, audio, and video.
Wayfair is personalizing customer experience, boosting operational efficiency, and addressing technical debt through AI.
FuriosaAI presented research papers on multimodal learning and diffusion model control at ECCV and CoLM.
Qwen has released Qwen2.5-VL, its next-generation vision-language model with agentic capabilities and long-video understanding.
In 2025, enterprise AI evolves toward industry specialization, RAG, and agentic AI.
Mistral AI has unveiled a major update to its AI assistant **le Chat**, adding web search, Canvas, and image generation features.
Kakao shares its AI technology application cases across six tracks, including AI agents, platforms, and multimodal AI.
Qwen2-VL, the latest vision-language model in the Qwen model series, has been released with enhanced video understanding and Agent capabilities.
OpenAI and Los Alamos National Laboratory have entered into a research partnership to evaluate the safety of using multimodal AI in bioscience research.
At Google Cloud Next 2024, an AI-based cloud tech stack centered on Gemini was unveiled.
Qwen has released Qwen-VL-Plus and Qwen-VL-Max, multimodal models with greatly enhanced image reasoning and recognition capabilities.
A total of 118 tech sessions and 8 keynotes, among other content, will be released on the official site and YouTube.
Team Naver increased multimodal model training throughput by 13.3% by resolving image token imbalance.
DeepSeek-V3 has been released, and API prices will be lowered starting February 8.