watsonx.ai hosts IBM Granite, Llama, Mistral and other models behind a governed API. Auth is IBM Cloud IAM: the apiKey is exchanged for a short-lived bearer token. Inference runs in a project (projectId) or a deployment space (spaceId); setting both is rejected. Prefer WatsonXConfig.fromValues over the primary constructor.
Value parameters
apiKey
IBM Cloud API key exchanged for IAM bearer tokens; redacted in toString.
apiKeyProvider
asked for the API key on every request instead of exchanging apiKey, so a rotated key is exchanged for a new IAM token without rebuilding the client (#2477): a section's apiKeyFile sets one.
apiVersion
watsonx API version date sent as the version query parameter, e.g. "2026-09-25".
baseUrl
watsonx.ai ML API base URL, e.g. "https://us-south.ml.cloud.ibm.com".
contextWindow
Model's total token capacity (prompt + completion combined).
iamUrl
IBM Cloud IAM token endpoint.
model
Model identifier, e.g. "ibm/granite-4-h-small".
projectId
watsonx project ID; may be empty only when spaceId is set.
reserveCompletion
Tokens held back from prompt history for the completion.
spaceId
Optional watsonx deployment space ID, used in place of the project (never together with it).
timeouts
how long a request and a stream may take: the section's timeouts block. An absent value keeps the client's own default (ProviderTimeouts).
The endpoint this config will contact, when it is known statically.
The endpoint this config will contact, when it is known statically.
Used by policy checks and diagnostics that need to know where traffic will go without knowing which provider it belongs to. None means the config carries no single URL — not that it makes no network calls.
The same config, its client asking provider for the key on each request, so a rotated key is used without rebuilding the client.
The same config, its client asking provider for the key on each request, so a rotated key is used without rebuilding the client.
The loader calls this for a section that sets apiKeyFile. The default returns this: a config whose client cannot take a provider ignores it, and its apiKeyProvider stays None - which the loader reports as an error for apiKeyFile and code setting a provider should check. A config that supports it overrides this with its own return type.
The same config with the given proxy and TLS settings.
The same config with the given proxy and TLS settings.
The loader calls this on every config a descriptor builds, as it does withTimeouts, so a descriptor does not read the http block itself. The default returns this: a config that does not carry the setting ignores it. A config that does overrides this with its own return type.
The loader calls this on every config a descriptor builds, passing the section's timeouts block, so a descriptor does not read the block itself. The default returns this: a config that does not carry timeouts, such as one supplied by another module, ignores them. A config that does overrides this with its own return type.