huawei-noah/bolt
Bolt is a deep learning library with high performance and heterogeneous flexibility.
Supports automated conversion from Caffe, ONNX, TensorFlow, and TFLite models with optimized inference across ARM (v7–v9), x86 (AVX2/AVX512), and mobile GPUs. Implements graph optimization, quantization down to 1-bit precision, and specialized acceleration for NLP workloads alongside traditional computer vision tasks. Includes on-device training capability and algorithm auto-tuning for minimal ROM/RAM footprint on embedded and edge devices.
957 stars. No commits in the last 6 months.
Stars
957
Forks
164
Language
C++
License
MIT
Category
Last pushed
Apr 11, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/huawei-noah/bolt"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
modern-fortran/neural-fortran
A parallel framework for deep learning
Cambridge-ICCS/FTorch
A library for directly calling PyTorch ML models from Fortran.
mlverse/torch
R Interface to Torch
Beliavsky/Fortran-code-on-GitHub
Directory of Fortran codes on GitHub, arranged by topic
BerkeleyLab/fiats
A deep learning library for use in high-performance computing applications in modern Fortran