LLM service implementation using NVIDIA’s NIM (NVIDIA Inference Microservice) API with OpenAI-compatible interface
NimLLMService
provides access to NVIDIA’s NIM language models through an OpenAI-compatible interface. It inherits from OpenAILLMService
and supports streaming responses, function calling, and context management, with special handling for NVIDIA’s incremental token reporting.
NVIDIA_API_KEY
.
OpenAILLMContextFrame
- Conversation context and historyLLMMessagesFrame
- Direct message listVisionImageRawFrame
- Images for vision processingLLMUpdateSettingsFrame
- Runtime parameter updatesLLMFullResponseStartFrame
/ LLMFullResponseEndFrame
- Response boundariesLLMTextFrame
- Streamed completion chunksFunctionCallInProgressFrame
/ FunctionCallResultFrame
- Function call lifecycleErrorFrame
- API or processing errors