66 articles
Researchers built a biomedical foundation model that combines vision and language pretraining with explicit medical concepts, aiming to make AI diagnoses more interpretable for doctors. The approach addresses a key limitation of current multimodal medical AI systems, which often lack transparency. Source: nature.com
AI Inspo has added Seedance 2.5 to its creative platform, enabling generation of 30-second AI videos from multimodal inputs with built-in audio. The update expands the tool's capabilities for longer, more complex AI-generated content. Source: somdnews.com
AI Inspo has added Seedance 2.5 to its creative platform, allowing users to generate AI videos up to 30 seconds long. The model supports multimodal inputs and comes with built-in audio generation. Source: streetinsider.com
A longitudinal study tested a multimodal LLM's ability to classify disease trajectories and stratify risk in patients with oral lichen planus. The research assessed how accurately the model could follow changes over time compared to traditional diagnostic methods. Source: nature.com
MiniMax M3 has launched with a 1-million-token context window, native multimodal capabilities, and pricing starting at $0.30 per million tokens. The model reportedly outperforms GPT-5.5 on the SWE-Bench Pro benchmark. Source: tech-insider.org
Pippit's new Seedance 2.5 update brings 4K video generation up to 30 seconds long, along with second-level timestamp precision and support for multiple input types. New features include multimodal references, audio-only prompts, 3D wireframe guidance, and improved tools for building narrative sequences. Source: manilatimes.net
A new comprehensive study examines preference alignment, a technique widely used to boost LLM performance, and explores how it specifically affects multimodal models. The research looks at both the benefits and limitations of applying this method beyond text-only systems. Source: google.com
Preference alignment has become key to improving performance of large language models, but its effects in multimodal settings remain less understood. A new comprehensive study examines how alignment techniques play out across different modalities. Source: machinelearning.apple.com
MiniMax H3 is an open-weight AI video generator built on multimodal reasoning, contextual regeneration, and an advanced architecture. Source: ca.news.yahoo.com
Black Forest Labs' new FLUX 3 model aims to replace the traditional patchwork of separate tools for image generation, video synthesis and other tasks with a single multimodal system. According to early comparisons, it outperforms rivals like Seedance 2.0 and Grok in several benchmarks. Source: sitepoint.com
MiniMax has released H3, a multimodal video generation model that reportedly tops independent benchmarks for video editing among all tracked models. However, the launch comes as the company faces a copyright lawsuit tied to its video generation technology. Source: techtimes.com
Google DeepMind researchers have introduced a new approach to prompt engineering that relies on visual cues rather than written instructions, claiming it delivers better results for guiding AI models. The method aims to bridge natural language processing with visual intelligence, pointing to a shift in how users interact with multimodal AI systems. Source: eu.36kr.com
A wave of multimodal AI progress coincides with Anduril reportedly raising $100 billion in new funding. The roundup also touches on politicians delivering AI-written speeches and other fresh AI developments shaking up the industry. Source: theneurondaily.com
Black Forest Labs has launched FLUX 3, a multimodal flow model capable of generating 20-second videos complete with native audio and action sequences. The model extends beyond image generation into video, sound, and robotics-related predictions. Source: marktechpost.com
Black Forest Labs released FLUX 3 on July 23, 2026, expanding beyond image generation into video, audio, and physical AI within a single model. The German company behind the FLUX image models is positioning this as a major step toward unified multimodal AI. Source: techtimes.com
Many AI apps end up juggling separate vendors for text, image, and video generation, which often happens gradually as products grow. GPTProto offers a unified API platform that lets developers integrate multiple AI model types into one application without managing several providers separately. Source: concept-phones.com
This week's roundup covers a wave of new AI model releases, including Claude Opus 5, Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, Qwen 3.8 Max, OpenAI Presence, FLUX 3, and MAI Image 2.5 Pro. The lineup reflects continued rapid competition among major AI labs across text, image, and multimodal models. Source: patmcguinness.substack.com
Black Forest Labs has released FLUX 3, expanding its FLUX family beyond still images to also generate up to 20-second videos with audio. The model is launching in a limited rollout rather than a full public release. Source: venturebeat.com
Black Forest Labs released Flux 3, a multimodal foundation model trained on images, video, and audio, capable of producing video clips with synchronized sound built in from the start. Source: the-decoder.com
The startup, valued at $3.25 billion, unveiled a new model that produces 20-second videos with matching audio and is already being tested on Audi's production lines. Source: cryptobriefing.com