Sources
Loading...
Additional media
Loading...
DeepNewz, mobile.
People-sourced. AI-powered. Unbiased News.
Transformers.js v3 Released: Bringing Power and Flexibility to Browser-Based Machine Learning https://t.co/o8GxlC45ZN #TransformersJS #MachineLearning #WebGPU #AIApplications #HuggingFace #ai #news #llm #ml #research #ainews #innovation #artificialintelligence #machinelearnin… https://t.co/vXqSpKbu6F
🧵 1/3 Transformers.js v3 Released: Bringing Power and Flexibility to Browser-Based Machine Learning Transformers.js v3, the latest release by Hugging Face, is a great step forward in making machine learning accessible directly within browsers. By leveraging the power of… https://t.co/fICzpDmOdq
Exciting updates in the latest release of xinference! 🌟 Enhanced transformers engine, dynamic batch support for FLUX.1, ML model quantization, and more. Faster response times and bug fixes included. Stay tuned for Python 3.12 support in v0.17.0! #xinference #AIUpdates
Hugging Face has announced the release of Transformers.js v3, marking a significant advancement in browser-based machine learning capabilities. This new version boasts WebGPU support, which can be up to 100 times faster than WASM, and introduces new quantization formats. Transformers.js v3 supports a total of 120 architectures and includes 25 new example projects and templates. The release also features over 1,200 pre-converted models, enhancing accessibility for developers and users alike. The update is part of a broader effort to democratize AI, with contributions from a diverse community of builders. The release has already seen impressive metrics, including 4.6 million model downloads and 2.5 million npm installs.
