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.
The demo processes five photos across three days to create a daily summary webpage that distinguishes between complete and incomplete meal records.
The new API processes text, images, audio, video, and PDFs to return confidence scores for up to 1,000 questions in a single pass at $0.10 per 1M input tokens.
The multimodal model processes text, images, audio, video, and PDFs to return Boolean probabilities, choices, and scores for policy checks and ranking.
if(kakao)26 highlighted the direction of transitioning from conversational AI to execution- and trust-centric agentic AI.
The model achieves 6 confirmed strengths across 11 categories, including a #1 rank in Humanity's Sixth Sense and #2 in SWE-Bench Pro V2.
The 1-trillion parameter native multimodal model is fully open-weight, trained in Europe, and offers 96.6% refusal rate for malicious cyber requests.
The new 8B and 70B models add native image and audio understanding, thinking mode, and a 4x larger context window under an Apache 2.0 license.
The update introduces opt-in 'danger zone' flags for image support, a new 'box' type, and extended context lengths, alongside a new model version focused on multilingual improvements.
AesCode achieves an overall score of 85.89 on infographic generation benchmarks, outperforming GPT-5.5 and Claude Opus 4.8 by using decoupled cross-modal rewards.
The study presents the first tri-modal Masked Diffusion Models pretrained on text, image-text, and audio-text data, achieving strong results at 3B scale.
Clef-omni handles audio and video inputs natively, while Clef-flash drops to $0.038 per million input tokens but reduces its hosted context window to 24k.
The 740M-parameter model unifies text, images, audio, and video into a shared embedding space under an Apache 2.0 license.
The model features 49 billion active parameters and will have its weights released by the end of October 2026.
Users can access the open-weight, multimodal model via a single API key without a separate Mistral account.
The framework improved GenEval scores from 0.747 to 0.808 on Qwen-image-2512 by using the model's own critiques as training signals.
Billing for AI Search begins November 1, 2026, with a free tier offering 5M ingestion tokens and 1,000 semantic queries per month.
The new sampler requires no additional training and uses one model forward pass per denoising step to revise low-confidence tokens.
The feature is available in Gemini Enterprise and supports seamless language switching across 97 languages.
inclusionAI has released Realtime-Venus, a 9B-scale multimodal dialogue model capable of processing video and audio in real time.
Qwen has released Qwen3.8-Omni-Flash, featuring enhanced real-time audio-visual interaction and agent capabilities, along with related open-source tools.
Meta has released the FLAT model, which integrates understanding and generation by mapping images and text into a single continuous token sequence.
Sierra has unveiled a multimodal AI agent that automatically switches between voice, text, and visual elements based on conversational context.
DeepSeek has released V4-Flash-Vision-Exp, significantly enhancing multimodal agent performance while maintaining text capabilities.
Box integrates Gemini Embeddings 2 to implement multimodal enterprise AI agents that understand diverse formats such as text, images, and tables.