Statements (54)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
software
|
| gptkbp:compatibleWith |
gptkb:ONNX
gptkb:PyTorch TensorFlow |
| gptkbp:developedBy |
NVIDIA
|
| gptkbp:features |
user authentication
data preprocessing health checks metrics and logging performance profiling role-based access control automatic scaling custom metrics model versioning logging and auditing request tracing model metadata dynamic batching |
| https://www.w3.org/2000/01/rdf-schema#label |
Triton Studio
|
| gptkbp:integratesWith |
gptkb:Grafana
gptkb:Prometheus Kubernetes |
| gptkbp:offers |
HTTP/gRPC APIs
custom backends |
| gptkbp:provides |
API documentation
load balancing model serving capabilities model optimization tools model lifecycle management model repository version control for models inference optimization secure inference |
| gptkbp:releasedIn |
2021
|
| gptkbp:supports |
gptkb:Java
gptkb:C++ Python NVIDIA GPUs A/B testing monitoring and alerting containerization cloud deployment multiple frameworks real-time inference batching edge deployment ensemble models ARM CPUs distributed inference multi-model serving offline inference x86 CPUs GPU_acceleration |
| gptkbp:usedFor |
machine learning inference
|