Services

Every service in the SDK follows the same builder pattern: configure it once with credentials, endpoint, and (where applicable) a model ID, then call its methods to interact with the API. Service instances are thread-safe and intended to be shared, so create one per service per application, not one per request.

Inference

Service	What it does
Chat	Conversational AI - synchronous and streaming, tool calling, vision, reasoning, structured output
Embedding	Convert text to dense vectors for semantic search, similarity, and RAG
Rerank	Score and sort a list of candidate passages against a query
Time Series	Forecast time series data using IBM Granite TTM models
Deployment	Target a deployed model by deploymentId for chat and forecasting
Model Gateway	Chat with third-party models (OpenAI, Anthropic, etc.) via a unified IBM-managed gateway
Foundation Model	Browse the model catalog - filter by provider, task, function, lifecycle

Document processing

These services read from and write to IBM Cloud Object Storage (COS). A CONNECTION_ID and BUCKET_NAME are required in addition to the standard credentials. See Setup & Prerequisites.

Service	What it does
Create Schema	Generate a key-value extraction schema from sample documents
Improve Schema	Refine an existing schema using additional examples
Merge Schema	Consolidate multiple schemas into one
Cluster Schema	Group a set of schemas into semantically similar clusters
Text Extraction	Extract structured key-value pairs from documents in COS
Text Classification	Classify documents stored in COS

Utilities

Service	What it does
Tool	Invoke IBM-hosted utility tools (search, weather, Python interpreter, RAG)
Tokenization	Count tokens and retrieve the individual token strings for a given model
Detection	Detect harmful content (HAP), PII, and safety violations
File	Upload, list, retrieve, and delete files used as batch job inputs
Batch	Submit high-volume asynchronous inference jobs from JSONL files