Warboy Product Page
High performance data center accelerator for advanced computer vision deployment on cloud and edge
Superior performance and versatility
Novel architecture and software enable fast and versatile computer vision deployment
Computer Vision Benchmarks
WARBOY vs. GPU Latencies (ms)
*Lower value is better
Measurements by FuriosaAI internally on current specifications and/or internal engineering calculations. System Configuration: [11th Gen Intel Core i9-11900K @ 3.50 GHz x 16, 64GB DDR4, 1x WARBOY or 1x Nvidia A2.
MLPerf
Image Classification (ResNet-50)
Performance/Price
MLPerf™v1.1 Inference Closed; SSD-Small: 1.1-071, 1.1-129, ResNet-50: 1.1-099, 1.1-129. Furiosa Warboy result is submitted in the Preview category, and Nvidia T4's SSD-Small and ResNet-50 results are submitted by Alibaba and Lenovo respectively in the Available category. Price is not the primary metric of MLPerf. The MLPerf name and logo are trademarks of MLCommons Association in the United States and other countries. All rights reserved. Unauthorized use strictly prohibited. See www.mlcommons.org for more information.
Object Detection (SSD - MobileNetV1)
Offline Throughput
Official MLPerf™ score of v2.0 Inference Edge: Closed. Retrieved from https://mlcommons.org/en/infer... 12 May 2022, entry 2.0-142. Price is the primary metric of MLPerf. The MLPerf™ name and logo are trademarks of MLCommons Association in the United States and other countries. All rights reserved. Unauthorized use strictly prohibited. See www.mlcommons.org for more information.
Image Classification (ResNet-50)
Object Detection (SSD-Small)
Single-Stream Performance (Image-per-second throughput)
MLPerf™v1.1 Inference Closed; SSD-Small: 1.1-071, 1.1-129, ResNet-50: 1.1-099, 1.1-129. Furiosa Gen 1 NPU result is submitted in the Preview category, and Nvidia T4's SSD-Small and ResNet-50 results are submitted by Alibaba and Lenovo respectively in the Available category. Price is not the primary metric of MLPerf. The MLPerf name and logo are trademarks of MLCommons Association in the United States and other countries. All rights reserved. Unauthorized use strictly prohibited. See www.mlcommons.org for more information.
Powered by Furiosa's Gen 1 NPU today
Intelligent video applications for smart cities, workplace safety, and quality control
Super Resolution applications for media and entertainment
Furiosa Model Zoo
Check out the full list of models that our Gen 1 NPU supports.
If you don’t see your model, our team will work with you to deploy the right solution.
Enterprise & Cloud-Ready Stack
Workflow Integration
Our platform integrates seamlessly with your existing workflow, offering easy-to-use APIs, support for popular AI frameworks like PyTorch, ONNX, TensorFlow Lite, and compatibility with NumPy data structures.
Model Optimization
Leveraging our advanced compiler, we ensure your models achieve peak performance per watt. With our profiling tools identifying performance bottlenecks, we can further enhance model performance and efficiency.
Flexible Deployment
Furiosa hardware provides the flexibility to choose multiple processing elements (PEs) depending on your workloads. It’s designed for data center adaptability, incorporating containerization and Kubernetes, to facilitate the swift scaling of AI projects.
