Muse Spark is a people-first AI model by Meta, built for everyday users and developers. It stands out with multimodal understanding, fast reasoning, and subagent workflows—delivering smarter answers, visual insights, and real-time, context-aware assistance.
Meta is accelerating its custom silicon roadmap with four new MTIA chips in two years. Built with an inference-first focus and native PyTorch integration, they are designed to cost-effectively power GenAI at a massive consumer scale.
SAM Audio is a unified model that separates any sound from any source. Use text ("dog barking"), visual clicks on video, or time spans to isolate specific audio. It unifies speech, music, and sound effect separation into one promptable model.
I'm not actually a user of these glasses, but I'm always interested in what's new, and I have to say that Meta is really putting a lot into this project.
For 2026, it comes with the following:
Glasses are not full AR, but Meta's first with a heads-up display
Oakley Meta Glasses are a new category of performance AI eyewear from Meta & Oakley. Capture hands-free video, listen to music with open-ear speakers, and get real-time answers from Meta AI. Preorders open July 11.
Meta's Omnilingual ASR is an open-source (Apache 2.0) speech recognition model supporting 1,600+ languages. It uses an LLM-based architecture that can be extended to new languages with just a few in-context examples, without retraining.
Meta introduces SAM 3 and SAM 3D. SAM 3 segments and tracks objects in images and videos using text or visual prompts. SAM 3D reconstructs 3D objects and bodies from single images. Try them now in the new Segment Anything Playground.
With the release of Llama 3.1 405B, TogetherCompute built LlamaCoder — an open source web app that can generate an entire app from a prompt. The repo has now been cloned by hundreds of devs on GitHub and starred 2K+ times.
Meta Ray-Ban Display is a new category of AI glasses featuring a private, in-lens color display. Paired with the Meta Neural Band allows for silent, intuitive control through subtle hand gestures, letting you interact with visual AI, messages, and more.
AudioCraft is Meta's simple framework that generates high-quality, realistic audio and music from text-based user inputs after training on raw audio signals as opposed to MIDI or piano rolls.
