#2828th of 56 in ML frameworks and libraries
transformers.js
Runs Hugging Face models in browser or Node via ONNX Runtime
- Stars
- 16.3k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- JavaScript
Missed the latest GitHub check; these facts may be older.
Overview
Transformers.js mirrors the Python transformers `pipeline` API in JavaScript and runs ONNX-converted models client-side with ONNX Runtime. It covers NLP, vision, audio and multimodal tasks such as text generation, speech recognition, text-to-speech, embeddings and object detection. It runs on CPU via WASM by default, with optional WebGPU and quantized dtypes (fp32, fp16, q8, q4).
Who it is for: JavaScript developers running ML models client-side in browsers or Node
Strengths
- Pipeline API closely matches Python transformers, easing code translation
- Runs in the browser with no server; models load from the Hugging Face Hub
- Optional WebGPU and q4/q8/fp16 dtypes to cut download size and speed up inference
- Wide task coverage: NLP, vision, audio, embeddings, document QA, background removal
Weaknesses
- Models must be in ONNX format; others need conversion with Optimum
- README calls WebGPU experimental in many browsers
- Unsupported tasks include text-to-image, visual QA, mask generation and tabular tasks
- Default setup fetches models and WASM binaries from remote hosts unless configured locally
What it needs
- GPU optional
- Needs ONNX Runtime, Hugging Face Hub (default model source)
- Models: ONNX, Hugging Face Hub models
Also in ML frameworks and libraries
See all 56| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 95 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 92 out of 100 |
| #4 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 89 out of 100 |
| #5 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 89 out of 100 |