Meta

Action100M v1.0

Action100M is a large-scale video action dataset developed by Meta, designed to advance research in action recognition and understanding. It comprises 100 million YouTube video clips, each annotated with hierarchical action labels, enabling the study of complex human activities in diverse contexts.

Llama 3.1

Meta's latest open-source large language model with 405B parameters. Offers state-of-the-art performance on benchmarks while being freely available for commercial use.

Muse Spark 1.1 v1.1

Muse Spark 1.1 is a multimodal reasoning model designed for agentic tasks, featuring enhanced capabilities in tool use, computer interaction, and complex coding. Developed by Meta Superintelligence Labs, it supports long-context management and efficient multi-agent orchestration for enterprise-grade workloads.

FFmpeg at Meta: Media Processing at Scale

Meta has integrated FFmpeg, an open-source multimedia framework, to enhance media processing capabilities across its platforms. By collaborating with FFmpeg developers, FFlabs, and VideoLAN, Meta has contributed to the development of features like threaded multi-lane transcoding and real-time quality metrics, which have been incorporated into the upstream FFmpeg project.

Perception Encoder Audiovisual (PE-AV) v1.0

Meta's Perception Encoder Audiovisual (PE-AV) is a state-of-the-art multimodal model that embeds audio, video, audio-video, and text into a joint embedding space. Trained using contrastive learning on approximately 100 million audio-video pairs with text captions, PE-AV enables powerful cross-modal retrieval and understanding across audio, video, and text modalities. ([marktechpost.com](https://www.marktechpost.com/2025/12/22/meta-ai-open-sourced-perception-encoder-audiovisual-pe-av-the-audiovisual-encoder-powering-sam-audio-and-large-scale-multimodal-retrieval/?utm_source=openai))

Muse Code vbeta

Muse Code is a terminal-based coding agent powered by the Muse Spark 1.2 model, designed to handle complex software engineering tasks across large repositories. It features persistent background agents, repository-scale execution, and built-in verification to automate planning, coding, and debugging workflows.