AI Rank
中文
← Open Source
visheratin

web-ai

Run modern deep learning models in the browser.

InfrastructureInference service optimizationTypeScript
Open on GitHub
Momentum
+0stars in 24 hours0.0%
848
Stars
43
Forks
+0
This week
4
Contributors
Created 2022-11-28 · Updated 2026-09-18 · #13244 today
Top developers
visheratinGarciaLnkmaccman

visheratin/web-ai is ranked 13,244 with 848 stars.

Same-category comparison

visheratin/web-ai is ranked 13,244 among these 8 peers; triton-inference-server/pytriton is ranked 13,232.

Same-category comparison
NameRankMetrics
jianshuo/ccglass13,204
Language:
JavaScript
Stars:
852
Forks:
48
jatinchowdhury18/RTNeural13,217
Language:
C++
Stars:
851
Forks:
91
Description:
Real-time neural network inferencing.
NickSwardh/YoloDotNet13,230
Language:
C#
Stars:
849
Forks:
149
Description:
YoloDotNet - A C# . NET 8.0 project for Classification, Object Detection, OBB Detection, Segmentation and Pose Estimation in both images and live video streams.
triton-inference-server/pytriton13,232
Language:
Python
Stars:
849
Forks:
63
Description:
PyTriton is a Flask/FastAPI-like interface that simplifies Triton's deployment in Python environments.
bentoml/Yatai13,309
Language:
TypeScript
Stars:
839
Forks:
76
Description:
Model Deployment at Scale on Kubernetes 🦄️.
scaleapi/llm-engine13,317
Language:
Python
Stars:
839
Forks:
87
Description:
Scale LLM Engine public repository.
RedisAI/redis-inference-optimization13,321
Language:
C
Stars:
838
Forks:
105
Description:
A Redis module for serving tensors and executing deep learning graphs.
UniversalRobots/Universal_Robots_ROS2_Driver13,323
Language:
C++
Stars:
838
Forks:
363
Description:
Universal Robots ROS2 driver supporting CB3 and e-Series.
Key parameters and specifications
Key parameters and specifications
ParameterValue
LanguageTypeScript
CategoryInfrastructure
SubcategoryInference service optimization
Stars848
Forks43
Contributors4
Created date2022-11-28
Updated date2026-09-18
Stars in 1 day0
Stars in 7 days0
Rank13,244
Previous rank13,132
Rank change-112
7-day stars rank13,838
Previous 7-day stars rank13,852
7-day stars rank change14
Total stars rank12,729
Previous total stars rank12,706
Total stars rank change-23
CountryUSA
README summary

Web AI Web AI is a TypeScript library that allows you to run modern deep learning models directly in the web browser or in Node. You can easily add AI capabilities to your web applications without the need for complex server-side infrastructure or third-party APIs. Create a model with one line of code, get the result with another one. Web AI runs the models using ONNX Runtime, which has rich support for of all kinds of operators. It

means that any model will work just fine. When using in the browser, Web AI stores the downloaded models in IndexedDB using localforage. You can configure the size of the cache dynamically. All heavy operations - model creation and inference - are offloaded to a separate thread so the UI does not freeze. More information about how to use the library can be found in the wiki. How to install The library can be installed via . js version: Status The library is

under active development. If something does not work correctly, please file an issue on GitHub. Sponsors Continuing work on this project is sponsored by Reflect - awesome app for taking notes. Thanks to AlgoveraAI for the grant under their AI project financing program.

Frequently asked questions

How many stars and forks are recorded?

visheratin/web-ai has 848 stars and 43 forks.

What is the repository rank?

visheratin/web-ai is ranked 13,244.

What language and category are recorded?

visheratin/web-ai has 848 stars, uses TypeScript and belongs to Infrastructure.

Which repository is closest in rank?

visheratin/web-ai is compared with triton-inference-server/pytriton, ranked 13,232 with 849 stars.

README

Web AI

Web AI is a TypeScript library that allows you to run modern deep learning models directly in the web browser or in Node.js. You can easily add AI capabilities to your web applications without the need for complex server-side infrastructure or third-party APIs.

Features:

  • Easy to use. Create a model with one line of code, get the result with another one.
  • Powered by ONNX runtime. Web AI runs the models using ONNX Runtime, which has rich support for of all kinds of operators. It means that any model will work just fine.
  • Built-in caching. When using in the browser, Web AI stores the downloaded models in IndexedDB using localforage. You can configure the size of the cache dynamically.
  • Web worker support. All heavy operations - model creation and inference - are offloaded to a separate thread so the UI does not freeze.

More information about how to use the library can be found in the wiki.

How to install

The library can be installed via npm.

Browser version:

npm install @visheratin/web-ai

Node.js version:

npm install @visheratin/web-ai-node

Status

The library is under active development. If something does not work correctly, please file an issue on GitHub. Contributions are very welcome.

Sponsors

  • Continuing work on this project is sponsored by Reflect - awesome app for taking notes.
  • Thanks to AlgoveraAI for the grant under their AI project financing program.
Nearby in the ranking
Similar items