aisdk (version 1.1.0)

NvidiaLanguageModel: NVIDIA Language Model Class

Description

Language model implementation for NVIDIA's chat completions API. Inherits from OpenAI model but adds support for NVIDIA-specific features like "enable_thinking" and reasoning content extraction.

Super classes

aisdk::LanguageModelV1 -> aisdk::OpenAILanguageModel -> NvidiaLanguageModel

Methods

Public methods

• NvidiaLanguageModel$parse_response()

• NvidiaLanguageModel$clone()

Inherited methods

• aisdk::LanguageModelV1$generate()
• aisdk::LanguageModelV1$has_capability()
• aisdk::LanguageModelV1$stream()
• aisdk::OpenAILanguageModel$build_payload()
• aisdk::OpenAILanguageModel$build_stream_payload()
• aisdk::OpenAILanguageModel$do_generate()
• aisdk::OpenAILanguageModel$do_stream()
• aisdk::OpenAILanguageModel$execute_request()
• aisdk::OpenAILanguageModel$format_tool_result()
• aisdk::OpenAILanguageModel$get_config()
• aisdk::OpenAILanguageModel$get_history_format()
• aisdk::OpenAILanguageModel$initialize()

---

Method parse_response()

Parse the API response into a GenerateResult. Overrides parent to extract NVIDIA-specific reasoning_content.

Usage

NvidiaLanguageModel$parse_response(response)

Arguments

response

The parsed API response.

Returns

A GenerateResult object.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

NvidiaLanguageModel$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.