snap-research/EfficientFormer

EfficientFormerV2 [ICCV 2023] & EfficientFormer [NeurIPs 2022]

44
/ 100
Emerging

Combines lightweight convolution blocks with pure transformer layers optimized via latency-driven architecture search, eliminating the need for hybrid CNN designs while achieving MobileNet-level speed. Supports end-to-end deployment pipelines to CoreML and ONNX formats with built-in profiling tools, enabling direct iPhone benchmarking and real-time inference on resource-constrained mobile hardware.

1,109 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 18 / 25

How are scores calculated?

Stars

1,109

Forks

94

Language

Python

License

Last pushed

Aug 13, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/snap-research/EfficientFormer"

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