transformers.js and bumblebee

Bumblebee appears to be a complementary tool to Transformers.js, likely providing a simplified API or additional utilities for running on-device models specifically within the browser environment, leveraging Transformers.js for the underlying inference engine.

transformers.js
68
Established
bumblebee
25
Experimental
Maintenance 23/25
Adoption 10/25
Maturity 16/25
Community 19/25
Maintenance 13/25
Adoption 3/25
Maturity 9/25
Community 0/25
Stars: 15,538
Forks: 1,104
Downloads:
Commits (30d): 38
Language: JavaScript
License: Apache-2.0
Stars: 3
Forks:
Downloads:
Commits (30d): 0
Language: TypeScript
License: MIT
No Package No Dependents
No Package No Dependents

About transformers.js

huggingface/transformers.js

State-of-the-art Machine Learning for the web. Run 🤗 Transformers directly in your browser, with no need for a server!

Leverages ONNX Runtime with WASM execution (CPU) or WebGPU (GPU) to run quantized transformer models efficiently in browsers, supporting optional 4-bit and 8-bit quantization. Provides a JavaScript API functionally equivalent to Hugging Face's Python library, enabling seamless conversion of PyTorch, TensorFlow, and JAX models via Optimum. Spans NLP, computer vision, audio, and multimodal tasks across text classification, object detection, speech recognition, and zero-shot learning.

About bumblebee

joshxfi/bumblebee

🐝 Run on-device models directly from your browser via Transformers.

Scores updated daily from GitHub, PyPI, and npm data. How scores work