在浏览器里直接跑 Hugging Face 的 Transformer 模型，不用后端服务器，前端就能做文本、视觉等机器学习推理。
Transformers.js 的官方示例合集，提供在浏览器和 Node.js 里直接跑预训练模型的演示与完整应用，方便你照着改出自己的前端 AI 功能。
…ncedb/pull/4890 🐛 Bug Fixes - fix(node): upgrade transformers.js to 4.3.2 to pick up patched sharp by @wjones127 in https://github.com/lancedb/lancedb/pull/4885 📚 Documentat
…3-VL by @xenova in https://github.com/huggingface/transformers.js/pull/1821 test: stabilize Whisper, Idefics3, and tokenizer CI timeouts by @nico-martin in https://github.com/
🚀 Transformers.js v4.3 — Structured Output, New Models, WebGPU upgrade, Documentation Overhaul This release adds structured output, three new model architectures, WebGPU supp
🚀 Transformers.js v4.2 — Tool calling, simpler internals, and privacy filtering - Added tools to TextGenerationPipeline in [ 1655](https://github.com/huggingface/transformers
🚀 Transformers.js v4.1 — Gemma 4, KV cache improvements, and new quantization dtypes - Add support for Gemma 4 in [ 1627](https://github.com/huggingface/transformers.js/pull/
…mples/benchmark/ demo that compares MLC (WebGPU), Transformers.js (ONNX/WASM), and Flare (GGUF/WASM) engines side-by-side. Measures model load time, TTFT, and tokens/sec with