mit-han-lab/hardware-aware-transformers

[ACL'20] HAT: Hardware-Aware Transformers for Efficient Natural Language Processing

45
/ 100
Emerging

This project helps machine translation practitioners deliver faster, more efficient language models on various devices. You input a machine translation task and the target hardware (like a Raspberry Pi or a powerful server CPU/GPU), and it outputs a specialized, pre-trained translation model optimized for speed and size on that hardware, without sacrificing translation quality. This is ideal for researchers and engineers deploying language models in diverse environments.

336 stars. No commits in the last 6 months.

Use this if you need to deploy high-quality machine translation models that run quickly and efficiently on specific hardware, especially devices with limited resources.

Not ideal if you are looking for a general-purpose, off-the-shelf translation service without specific hardware optimization needs.

machine-translation natural-language-processing edge-ai model-optimization efficient-inference
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

336

Forks

50

Language

Python

License

Last pushed

Jul 14, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/mit-han-lab/hardware-aware-transformers"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.