3 articles tagged with "Multimodal AI"
NVIDIA released Nemotron 3.5 Content Safety, a 4B-parameter model that combines multimodal input evaluation, multilingual reach across 140 languages, custom enterprise policy enforcement, and auditable reasoning traces in one inference call. The model addresses critical gaps in production AI safety pipelines.
Google unveiled Gemini Omni, a multimodal model that generates and edits video through natural language, alongside Gemini 3.5 Flash, designed for complex agentic workflows. Both models are rolling out to consumers and developers with significant implications for content creation and enterprise automation.
OpenAI has released GPT-5.5, marking a significant iteration in large language model capabilities. The model introduces enhanced multimodal reasoning, extended context windows, and improved computational efficiency without requiring architectural overhaul.
