AI Briefing#multimodal
The latest AI and developer news about #multimodal, with the original source and a short summary.
The latest AI and developer news about #multimodal, with the original source and a short summary.
Thinking Machines Lab has released a real-time interaction model for human-AI collaboration.
llama.cpp is undergoing updates to resolve conflicts between MTP and multimodal features and support their integration.
Interfaze has unveiled a new model architecture and API for OCR, STT, and structured output.
Google has made Gemini 3.1 Flash-Lite generally available.
Gemini API File Search now supports image-text RAG and page citations.
Supports real-time conversation and high-quality TTS simultaneously through OpenAI API compatibility and a process isolation architecture.
By jointly training the tokenizer and the generative model, the approach achieved FID 1.48 on ImageNet 256x256.
Tuna-2 outperformed Tuna-R and the original Tuna using a pixel embedding approach that removes the vision encoder.
Signs have emerged that Google is testing a new Omni model in the Gemini video generation tab.
GLM-5V-Turbo is an agent model that integrates multimodal perception, tool use, and end-to-end verification.
Using the UEyes dataset, VLM gaze predictions were compared against real eye-tracking data.
SenseNova-U1 has been unveiled as a native multimodal model that unifies understanding, reasoning, and generation.
NVIDIA has unveiled the commercially usable multimodal Nemotron 3 Nano Omni.
Nvidia has unveiled Nemotron 3 Nano Omni, an open-weight multimodal model.
NVIDIA has unveiled Nemotron 3 Nano Omni, a multimodal model that handles documents, audio, and video.
NVIDIA has unveiled the multimodal Nemotron Nano Omni 30B A3B Reasoning.
NVIDIA has unveiled Nemotron 3 Nano Omni, which integrates vision, audio, and language.
DeepSeek has teased the release of a new vision model.
Pocket LLM v1.5.0 has been released, supporting voice and image input as well as OCR.
Pocket LLM v1.5.0 has been released, adding voice and image input, OCR, and camera capture to Android offline chat.
ChatGPT Images 2.0 added text to an image that wasn't in the prompt.
Introducing Flipbook, an experimental interface where an image model directly renders pixels to generate web pages in real time.
A multimodal prompt injection dataset v5 with 503,358 samples has been released.
OpenAI is privately testing GPT Image 2, with a launch believed to be imminent.