Generative CS 0.8.2

Full Changelog: v0.8.1...v0.8.2

Generative CS 0.8.1

Full Changelog: v0.8.0...v0.8.1

Generative CS 0.8.0

Full Changelog: v0.7.0...v0.8.0

Generative CS 0.7.0

• Added constants for new OpenAI chat completion models: O1Preview, O1Preview20240912, O1Mini, and O1Mini20240912.
• Added constants for new OpenAI moderation model: OmniModerationLatest and OmniModeration20240926.
• Added new IsIllicit, IllicitScore, IsIllicitViolent, and IllicitViolentScore properties to the ModerationResult record.
• Changed default moderation model to omni-moderation-latest.

Full Changelog: v0.6.2...v0.7.0

Generative CS 0.6.2

• Added IsParallelFunctionCallingOn property to OpenAI ChatCompletionOptions (equivalent of parallel_tool_calls).

Full Changelog: v0.6.1...v0.6.2

Generative CS 0.6.1

Full Changelog: v0.6.0...v0.6.1

Generative CS 0.6.0

• The API key can now be configured per-request by providing it in options (for example ChatCompletionOptions).
• Removed apiKey parameter from various client constructors.

Full Changelog: v0.5.0...v0.6.0

Generative CS 0.5.0

• Added support for dynamic system messages that can change after the chat has started. To take advantage of it, set SystemMessageCallback property of ChatCompletionOptions instead of adding the system message to the conversation.
• Added CompleteAsync method overload with systemMessage and userMessage parameters (OpenAI client only).

Full Changelog: v0.4.1...v0.5.0

Generative CS 0.4.1

• Added constant for gpt-4o-2024-08-06 chat completion model: ChatCompletionModels.OpenAI.GPT4o20240806.

Full Changelog: v0.4.0...v0.4.1

Generative CS 0.4.0

• Added token usage tracking. To use this feature, create an instance of TokenUsageTracker and pass it as an argument to CompleteAsync,StreamCompletionAsync, GetEmbeddingAsync or GetBase64EmbeddingAsync. Currently only OpenAI client is supported.

Full Changelog: v0.3.6...v0.4.0