For the complete documentation index, see llms.txt. Markdown versions of all docs pages are available by appending .md to any docs URL.
TypeSafe
Verified : Code examples on this page have been automatically tested and verified.Route agentgateway LLM traffic to TypeSafe Jev with the built-in provider preset.
Use TypeSafe Jev as an LLM provider in agentgateway. The typesafe preset uses the TypeSafe API and the SystemOne provider format for OpenAI Decisions API requests.
Configuration
Review the following example configuration.# yaml-language-server: $schema=https://agentgateway.dev/schema/config
llm:
models:
- name: jev-latest
visibility: internal
provider: typesafe
params:
apiKey: "$TYPESAFE_API_KEY"
virtualModels:
- name: guardrail
routing:
weighted:
targets:
- model: jev-latest
weight: 100| Setting | Description |
|---|---|
name | The model name that matches incoming requests or virtual model targets. |
visibility | Set to internal when clients must not call the TypeSafe model directly. |
provider | The LLM provider, set to typesafe. |
params.apiKey | The TypeSafe API key. You can reference environment variables using the $VAR_NAME syntax. |
params.baseUrl | Optional. Use this setting to override the provider base URL. The default is https://api.typesafe.ai/v1. |
virtualModels[].routing.weighted.targets[].model | The TypeSafe model that the virtual model targets. |
Next steps
After you run agentgateway with the configuration from the Configuration section, you can send OpenAI Decisions API requests to the virtual model. The configured TypeSafe provider translates the requests to TypeSafe’s native /v1/systemone API before the requests reach TypeSafe. For a full guardrail workflow that uses this provider, see Jev.
Was this page helpful?
