Tencent Cloud TI-ONE
Managed machine learning platform for data preparation, model training, evaluation, deployment, and online inference on CPU/GPU resources.
At a Glance
About Tencent Cloud TI-ONE
Tencent Cloud TI-ONE is a managed machine learning and large-model training and inference platform for AI engineers. It covers data preparation, model building, training, evaluation, model management, and deployment.
Development and training
The platform provides automated modeling and hyperparameter tuning, an interactive development environment supporting Python and R, and support for PySpark, Spark, PyTorch, and TensorFlow. It supports CPU/GPU resource scheduling, training monitoring, and Git-based code management.
Data and deployment
Teams can manage and annotate text, image, and tabular datasets, including imported image annotations in Pascal VOC and COCO formats. Online model services support heterogeneous compute, elastic scaling, traffic allocation, and rolling updates.
Billing
Official documentation separates compute charges from TI-ONE software subscription charges. Resources bought directly through TI-ONE use pay-as-you-go billing. Existing CVM resources can be added to resource groups, with software billed according to the corresponding subscription or usage model. Costs depend on the selected resources; no single flat product price is represented here.
The original /products/ti URL is unavailable. This listing uses the verified official TI-ONE product page and product-specific documentation.
Community Discussions
Be the first to start a conversation about Tencent Cloud TI-ONE
Share your experience with Tencent Cloud TI-ONE, ask questions, or help others learn from your insights.
Pricing
Resource-based billing
Compute plus TI-ONE software subscription charges. Direct platform purchases use pay-as-you-go; existing CVM resources may use subscription or usage billing. Price varies by resources and region; USD is a catalog field, not a verified quote denomination.
- CPU/GPU compute resources
- TI-ONE software platform
Capabilities
Key Features
- Data preparation and annotation
- Automated modeling and hyperparameter tuning
- CPU/GPU training resource scheduling
- Interactive Python and R development
- Model evaluation and management
- Elastic online inference with traffic allocation and rolling updates
