Skip to main content

LLM / AI Backend

By default OpsWorker runs on its own managed AWS Bedrock backend, so most teams need no LLM configuration at all. There are two supported options:

  • AWS Bedrock (default): the managed backend, no setup required.
  • Custom Reasoning Model (Anthropic only): point a cluster at your own Anthropic API key from the portal.

For other model hosting (for example dedicated or private deployments), contact the OpsWorker team. There is no self-service Azure OpenAI or generic OpenAI-compatible option today.

🟠 AWS Bedrock

The default managed backend. OpsWorker uses Claude models for reasoning and Amazon Nova for routing, extraction, and summarization. No setup required.

Read more →

⚙️ Custom Reasoning Model

Per-cluster integration: point OpsWorker at your own Anthropic API key from the portal. Anthropic only, configured per cluster.

Read more →

🔵 Azure OpenAI

Not available as self-service today. See the supported options above, or contact OpsWorker for other model hosting.

Read more →

🧠 Bring Your Own LLM

Not available as self-service today. AWS Bedrock and the Anthropic Custom Reasoning Model are the supported options; contact OpsWorker for other hosting.

Read more →