Codeunit 7765 AOAI Text Completion Params
src/System Application/App/AI/src/Azure OpenAI/Text Completion/AOAITextCompletionParams.Codeunit.al149 lines, Copyright (c) Microsoft Corporation. MIT
#if not CLEAN29// ------------------------------------------------------------------------------------------------// Copyright (c) Microsoft Corporation. All rights reserved.// Licensed under the MIT License. See License.txt in the project root for license information.// ------------------------------------------------------------------------------------------------namespace System.AI;/// <summary>/// Represents the Completion parameters used by the API./// See more details at https://aka.ms/AAlsi39./// </summary>codeunit 7765 "AOAI Text Completion Params"{ Access = Public; InherentEntitlements = X; InherentPermissions = X; ObsoleteState = Pending; ObsoleteReason = 'Text completion models are retired by Azure OpenAI. Use GenerateChatCompletion instead.'; ObsoleteTag = '29.0'; var AOAITextCompletionParamsImpl: Codeunit "AOAI TextCompletionParams Impl"; /// <summary> /// Get the maximum number of tokens to generate in the completion. /// </summary> /// <returns>The maximum number of tokens to generate in the completion.</returns> /// <remarks>0 or less uses the API default.</remarks> procedure GetMaxTokens(): Integer begin exit(AOAITextCompletionParamsImpl.GetMaxTokens()); end; /// <summary> /// Get the sampling temperature to use. /// </summary> /// <returns>The sampling temperature to use.</returns> procedure GetTemperature(): Decimal begin exit(AOAITextCompletionParamsImpl.GetTemperature()); end; /// <summary> /// Get the nucleus sampling to use. /// </summary> /// <returns>The nucleus sampling to use.</returns> procedure GetTopP(): Decimal begin exit(AOAITextCompletionParamsImpl.GetTopP()); end; /// <summary> /// Get the suffix that comes after a completion of inserted text. /// </summary> /// <returns>The suffix that comes after a completion of inserted text.</returns> procedure GetSuffix(): Text begin exit(AOAITextCompletionParamsImpl.GetSuffix()); end; /// <summary> /// Get the presence penalty to use. /// </summary> /// <returns>The presence penalty to use.</returns> procedure GetPresencePenalty(): Decimal begin exit(AOAITextCompletionParamsImpl.GetPresencePenalty()); end; /// <summary> /// Get the frequency penalty to use. /// </summary> /// <returns>The frequency penalty to use.</returns> procedure GetFrequencyPenalty(): Decimal begin exit(AOAITextCompletionParamsImpl.GetFrequencyPenalty()); end; /// <summary> /// The maximum number of tokens to generate in the completion. The token count of your prompt plus max_tokens can't exceed the model's context length. Most models have a context length of 2048 tokens (except for the newest models, which support 4096). /// </summary> /// <param name="NewMaxTokens">The new maximum number of tokens to generate in the completion.</param> /// <remarks>If the prompt's tokens + max_tokens exceeds the model's context length, the generate request will return an error.</remarks> /// <remarks>The default value is 256 however a value of 0 or less will use the deployment model's default.</remarks> procedure SetMaxTokens(NewMaxTokens: Integer) begin AOAITextCompletionParamsImpl.SetMaxTokens(NewMaxTokens); end; /// <summary> /// Sets the sampling temperature to use, between 0 and 2. A higher temperature increases the likelihood that the next most probable token will not be selected. When requesting structured data, set the temperature to 0. For human sounding speech, 0.7 is a typical value /// </summary> /// <param name="NewTemperature">The new sampling temperature to use.</param> /// <error>Temperature must be between 0.0 and 2.0</error> procedure SetTemperature(NewTemperature: Decimal) begin AOAITextCompletionParamsImpl.SetTemperature(NewTemperature); end; /// <summary> /// An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both. /// </summary> /// <param name="NewTopP">New nucleus sampling to use</param> procedure SetTopP(NewTopP: Decimal) begin AOAITextCompletionParamsImpl.SetTopP(NewTopP); end; /// <summary> /// The suffix that comes after a completion of inserted text. /// </summary> /// <param name="NewSuffix">The new suffix that comes after a completion of inserted text.</param> procedure SetSuffix(NewSuffix: Text) begin AOAITextCompletionParamsImpl.SetSuffix(NewSuffix); end; /// <summary> /// Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. /// </summary> /// <param name="NewPresencePenalty">The new presence penalty to use.</param> /// <error>Presence penalty must be between -2.0 and 2.0</error> procedure SetPresencePenalty(NewPresencePenalty: Decimal) begin AOAITextCompletionParamsImpl.SetPresencePenalty(NewPresencePenalty); end; /// <summary> /// Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. /// </summary> /// <param name="NewFrequencyPenalty">The new frequency penalty to use.</param> /// <error>Frequency penalty must be between -2.0 and 2.0</error> procedure SetFrequencyPenalty(NewFrequencyPenalty: Decimal) begin AOAITextCompletionParamsImpl.SetFrequencyPenalty(NewFrequencyPenalty); end; /// <summary> /// Add the completion parameters to the payload. /// </summary> /// <param name="Payload">The JsonObject payload to add the completion parameters to.</param> [NonDebuggable] internal procedure AddCompletionsParametersToPayload(var Payload: JsonObject) begin AOAITextCompletionParamsImpl.AddCompletionsParametersToPayload(Payload); end;}#endif