Skip to main content
The Infercom Developer guide and API reference provide the tools you need to build applications using Infercom as an inference service.

​
API endpoints

​
Text generation

EndpointDescriptionBest for
/v1/chat/completionsOpenAI-compatible chat APIConversational applications, OpenAI SDK users
/v1/messagesAnthropic-compatible Messages APIClaude Code, LangChain Anthropic, Anthropic SDK users
/v1/responsesStructured output API for agentic workflowsCoding agents, tool calling, multi-step reasoning
See OpenAI compatibility for the Chat Completions API, Anthropic compatibility for the Messages API, or Responses API for agentic workflows.

​
Embeddings

EndpointDescriptionBest for
/v1/embeddingsGenerate vector embeddingsRAG, semantic search, classification
See Embeddings for usage examples.

​
Audio

EndpointDescriptionBest for
/v1/audio/transcriptionsTranscribe audio to textSpeech-to-text, meeting transcription
/v1/audio/translationsTranslate audio to EnglishMultilingual audio processing
See Audio for usage examples.

​
Infercom Inference Service

Code examples in this API reference include the specific URLs for developing on Infercom Inference Service.

​
Model sovereignty metadata

The /v1/models endpoint supports a ?verbose=true query parameter that returns detailed metadata for each model, including sovereignty information. Use the sn_metadata.region field to determine where a model is hosted:
  • "EU" — the model runs on Infercom’s EU infrastructure in Germany with full data sovereignty
  • Empty or absent — the model is available via SambaNova’s global infrastructure
See Supported models for details and code examples.