Tag

Magenta RealTime 2

A toolkit and open-weights system for real-time streaming music generation — offers two model sizes (230M / 2.4B), a Python inference library (JAX/MLX), and a C++ engine optimized for Apple Silicon for embedding into DAWs and apps; real-time streaming requires M‑series chips.

LangExtract

Extracts structured data from unstructured text with LLMs, mapping every extraction to its exact character span in the source for visual review. Uses few-shot examples, schema enforcement, and multi-pass chunking to handle long documents.

Google Analytics MCP Server

Exposes Google Analytics Admin and Data APIs as a local Model Context Protocol (MCP) server so LLMs can query accounts, run reports, funnels and realtime queries via standardized MCP tools. Intended for local prototypes and developer integrations; requires Google Cloud credentials.

A2UI: Agent-to-User Interface

Lets AI agents describe interactive UIs as declarative JSON instead of executable code; client apps render the components with native widgets from a pre-approved catalog, keeping agent-generated UI safe across trust boundaries.

Video models are zero-shot learners and reasoners

Argues a single web-scale generative video model handles vision tasks zero-shot the way LLMs handle language. Probes Veo 3 on segmentation, edge detection, image editing, physical and affordance reasoning, and puzzles like maze solving and symmetry.

Google Antigravity

Agent-first development platform: spawn autonomous agents that plan, edit, run terminals, and drive a browser to verify their own work, returning reviewable artifacts like plans and screenshots. Defaults to Gemini 3 Pro; also runs Claude and GPT-OSS.

Reverse-Engineering SynthID

Detects and surgically removes Google's SynthID watermark from images using multi-resolution spectral analysis and a resolution-aware codebook; provides a V3 bypass with high PSNR and strong phase-coherence reduction. Research-focused and intended for analysis/defense, not misrepresentation.

Stitch Agent Skills

Installs modular Agent Skills for the Stitch MCP server to expose Stitch-powered design and coding capabilities to coding agents. Includes CLI discovery/installation, UI-to-React conversion, design-system synthesis, multi-page site generation, and Remotion video export.

Waxal NLP Datasets

Provides open ASR and TTS speech data for 24 Sub‑Saharan African languages to train and evaluate speech models. Includes ~1,250 hours of transcribed ASR and ~235 hours of single‑speaker TTS with train/validation/test/unlabeled splits and mixed CC-BY licenses.

google/gemma-4-E4B-it

An instruction‑tuned Gemma 4 E4B multimodal model on Hugging Face that accepts text, images and audio and generates text; notable for 128K long context support, built-in thinking mode, and an on‑device‑friendly E4B architecture under an Apache‑2.0 license.

google/gemma-4-31B-it

Instruction-tuned Gemma 4 31B multimodal model that generates text from text+image inputs with up to 256K context. Dense 31B variant optimized for vision-language understanding, long-context reasoning, and coding; Apache‑2.0 licensed.

Gemma 4 26B A4B (google/gemma-4-26B-A4B-it)

Instruction-tuned Mixture-of-Experts multimodal model that generates text from text+image inputs while activating a 4B subset of parameters for faster inference; supports a 256K context window, multilingual vision-language tasks, and is available under Apache-2.0.