OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Transformers Js

Transformers Js is a JavaScript library and model-conversion tool for running supported Hugging Face transformer models in a web browser using ONNX Runtime for Web and a JavaScript tokenizer.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Transformers Js — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a web application needs browser-side inference with a supported T5 model and its developer can convert and serve the ONNX model files.

Who it is for

  • Web developers embedding browser-side model inference in a website.
  • Developers preparing T5 models for browser use.
  • Developers building browser translation or text-generation features.

Documented uses

  • Convert a supported Hugging Face T5 model to ONNX format.
  • Load a tokenizer and ONNX model in JavaScript.
  • Encode a task prompt, generate output tokens and decode them.
  • Build browser-side translation or text-generation features.
Before you choose

Requirements & limits

The repository is public and licensed under MIT.

  • Current documented model support is limited to T5; GPT-2, RoBERTa and InCoder are mentioned only as potential future additions.
  • Models must be converted to ONNX before the documented browser-loading workflow.
  • The library uses ONNX Runtime for Web and its own JavaScript-only tokenizer implementation.
  • The README documents greedy and top-k sampling.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Convert a T5 model for browser use

Prepare a Hugging Face T5 model in ONNX format for loading by the browser library.

Before you start

  • The project repository, including its conversion tool.
  • Python 3.
  • A model identifier and output directory.

Steps

  1. Run the documented converter command: python3 tools/convert_model.py t5-small models.

  2. Use the generated ONNX model files as the model files to serve for browser use.

Expected outcome

ONNX model files prepared for the browser library’s model-loading workflow.

Guide limitations

  • The documented supported model family is T5.
  • The guide covers conversion and serving preparation, not a verified deployment or execution test.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
praeclarum/transformers-js · Official homepage title · retained in its source language
How we review sourcesSuggest a correction