Build with Baseten
Baseten is a platform for deploying and serving AI models performantly,
scalably, and cost-efficiently.
Choose from common AI/ML usecases and modalities to get started on Baseten quickly.
Baseten makes it easy to deploy, serve, and scale AI models so you can focus on building, not infrastructure.
**Baseten is an infrastructure platform for AI/ML models that lets you:**
* [Package any model for production](/development/model/overview): Define dependencies, hardware, and custom code without needing to learn Docker. Build with your preferred frameworks (e.g. **PyTorch**, **transformers**, **diffusers**), [inference engines](development/model/performance/concepts) (e.g. **TensorRT-LLM**, **SGLang**, **vLLM**), and serving tools (e.g. **Triton**) as well as [any package](/development/model/configuration) installable via `pip` or `apt`.
* [Build complex AI systems](/development/chain/overview): Orchestrate multi-step workflows with [Chains](/development/chain/overview), combining models, business logic, and external APIs.
* [Deploy with confidence](/deployment/concepts): [Autoscale](/deployment/autoscaling) models, manage [environments](/deployment/environments), and roll out updates with zero-downtime deployments.
* [Run high-performance inference](/inference/concepts): Serve [synchronous](/inference/calling-your-model), [asynchronous](/inference/async), and [streaming](/inference/streaming) predictions with low-latency execution controls.
* [Monitor and optimize in production](/observability/metrics): Monitor performance, debug failures, and [export metrics](/observability/export-metrics/overview) with built-in observability tooling.
Resources
From deploying LLMs to optimizing inference and scaling ML models.
Prebuilt, ready to deploy in one click models like DeepSeek, Llama, and
Qwen.
API reference for calling deployed models, chains or managing models and
your workspace.
# Quick start
Source: https://docs.baseten.co/quickstart