+
+
+
+
+ Compute the similarity of two embeddings using cosine similarity.
Compute the similarity of two embeddings using cosine similarity.
Transcribes audio into the input language.
Translates audio into English.
Creates a completion for the chat message.
Stream variant of chatCompletion.
Creates a completion for the chat message.
Stream variant of chatCompletion.
Defines a closeable resource. This will be replaced by AutoCloseable once it becomes stable.
This is the main endpoint of the API. Returns the predicted completion for the given prompt, and can also return the probabilities of alternative tokens at each position if requested.
Stream variant of completion.
Given a prompt, the model will return one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
This is the main endpoint of the API. Returns the predicted completion for the given prompt, and can also return the probabilities of alternative tokens at each position if requested.
Stream variant of completion.
Edits is deprecated. Chat completions is the recommend replacement.
Creates a new edit for the provided input, instruction, and parameters.
Creates an embedding vector representing the input text.
Get a vector representation of a given input that can be easily consumed by machine learning models and algorithms.
Upload a file that contains document(s) to be used across various endpoints/features. Currently, the size of all the files uploaded by one organization can be up to 1 GB.
Returns information about a specific file.
Use FineTuning instead.
Immediately cancel a fine-tune job.
Use FineTuning instead.
Get fine-grained status updates for fine-tune job.
Use FineTuning instead.
Get fine-grained status updates for fine-tune job.
Use FineTuning instead.
Creates a job that fine-tunes a specified model from a given dataset.
Response includes details of the enqueued job including job status and the name of the fine-tuned models once complete.
Use FineTuning instead.
Gets info about to fine-tune job.
Immediately cancel a fine-tune job.
Gets info about to fine-tune job.
Creates a job that fine-tunes a specified model from a given dataset.
Get fine-grained status updates for fine-tune job.
Get fine-grained status updates for fine-tune job.
Immediately cancel a fine-tune job.
The ID of the fine-tuning job to cancel.
Get status updates for a fine-tuning job.
The ID of the fine-tuning job to get events for.
Identifier for the last event from the previous pagination request.
Number of events to retrieve.
Creates a job that fine-tunes a specified model from a given dataset.
Response includes details of the enqueued job including job status and the name of the fine-tuned models once complete.
Get info about a fine-tuning job.
The ID of the fine-tuning job.
List your organization's fine-tuning jobs.
Identifier for the last job from the previous pagination request.
Number of fine-tuning jobs to retrieve.
Manage fine-tuning jobs to tailor a model to your specific training data.
Immediately cancel a fine-tune job.
Get status updates for a fine-tuning job.
Get info about a fine-tuning job.
Creates a job that fine-tunes a specified model from a given dataset.
List your organization's fine-tuning jobs.
Creates an image given a prompt. Get images as base 64 JSON.
Creates an edited or extended image given an original image and a prompt. Get images as base 64 JSON.
Creates a variation of a given image. Get images as base 64 JSON.
Creates an image given a prompt. Get images as URLs.
Creates an edited or extended image given an original image and a prompt. Get images as URLs.
Creates a variation of a given image. Get images as URLs.
Creates an image given a prompt. Get images as base 64 JSON.
Creates an edited or extended image given an original image and a prompt. Get images as base 64 JSON.
Creates a variation of a given image. Get images as base 64 JSON.
Creates an image given a prompt. Get images as URLs.
Creates an edited or extended image given an original image and a prompt. Get images as URLs.
Creates a variation of a given image. Get images as URLs.
Defines the configuration parameters for logging.
Given an input text, outputs if the model classifies it as violating OpenAI's content policy.
Classifies if text violates OpenAI's Content Policy.
OpenAI Token
client logging configuration
http client timeout
extra http headers
OpenAI organization ID
OpenAI host configuration.
HTTP proxy url
rate limit retry configuration
additional custom client configuration
Use primary constructor with LoggingConfig instead.
import com.aallam.openai.api.logging.Logger
import com.openai.config.LoggingConfig
OpenAIConfig(token, LoggingConfig(logLevel, logger), timeout, organization, headers, host, proxy, retry)additional custom client configuration
OpenAI client configuration.
OpenAI Token
client logging configuration
http client timeout
extra http headers
OpenAI organization ID
OpenAI host configuration.
HTTP proxy url
rate limit retry configuration
additional custom client configuration
A pre-configured instance of OpenAIHost with the base URL set as https://api.openai.com/v1/.
Creates an instance of OpenAIHost configured for Azure hosting with the given resource name, deployment ID, and API version.
The name of your Azure OpenAI Resource.
The name of your model deployment.
The API version to use for this operation. This parameter should follow the YYYY-MM-DD format.
A pre-configured instance of OpenAIHost with the base URL set as https://api.openai.com/v1/.
Creates an instance of OpenAIHost configured for Azure hosting with the given resource name, deployment ID, and API version.
A class to configure the OpenAI host. It provides a mechanism to customize the base URL and additional query parameters used in OpenAI API requests.
Additional query parameters to be appended to all API requests to OpenAI. These can be used to provide additional configuration or context for the API requests.
Creates an instance of OpenAI.
secret API key
client logging configuration
http client timeout
extra http headers
OpenAI organization ID
OpenAI host configuration.
HTTP proxy url
rate limit retry configuration
additional custom client configuration
Creates an instance of OpenAI.
client config
OpenAI API.
Immediately cancel a fine-tune job.
Creates a completion for the chat message.
Stream variant of chatCompletion.
This is the main endpoint of the API. Returns the predicted completion for the given prompt, and can also return the probabilities of alternative tokens at each position if requested.
Stream variant of completion.
Creates a new edit for the provided input, instruction, and parameters.
Creates an embedding vector representing the input text.
Returns information about a specific file.
Upload a file that contains document(s) to be used across various endpoints/features. Currently, the size of all the files uploaded by one organization can be up to 1 GB.
Gets info about to fine-tune job.
Creates a job that fine-tunes a specified model from a given dataset.
Get fine-grained status updates for fine-tune job.
Get fine-grained status updates for fine-tune job.
Get status updates for a fine-tuning job.
Get info about a fine-tuning job.
Creates a job that fine-tunes a specified model from a given dataset.
List your organization's fine-tuning jobs.
Creates an image given a prompt. Get images as base 64 JSON.
Creates an edited or extended image given an original image and a prompt. Get images as base 64 JSON.
Creates a variation of a given image. Get images as base 64 JSON.
Creates an image given a prompt. Get images as URLs.
Creates an edited or extended image given an original image and a prompt. Get images as URLs.
Creates a variation of a given image. Get images as URLs.
Classifies if text violates OpenAI's Content Policy.
Transcribes audio into the input language.
Translates audio into English.
Specifies the retry strategy
the maximum amount of retries to perform for a request
retry base value
max retry delay
Given a prompt, the model will return one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
Get a vector representation of a given input that can be easily consumed by machine learning models and algorithms.
Manage fine-tuning jobs to tailor a model to your specific training data.
Defines the configuration parameters for logging.
Given an input text, outputs if the model classifies it as violating OpenAI's content policy.
OpenAI API.
OpenAI client configuration.
A class to configure the OpenAI host. It provides a mechanism to customize the base URL and additional query parameters used in OpenAI API requests.
Proxy configuration.
Specifies the retry strategy
Creates an instance of OpenAI.
The audio file to transcribe, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
Builder of TranscriptionRequest instances.
The audio file to transcribe, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
Builder of TranscriptionRequest instances.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
The audio file to transcribe, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
Request to transcribe audio into the input language.
The audio file to transcribe, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
Create transcription response.
text format depends on TranscriptionRequest's responseFormat. Remaining field are provided only in case of response format verbose_json.
Transcription text. The format depends on TranscriptionRequest's responseFormat.
Transcription text. The format depends on TranscriptionRequest's responseFormat.
The audio file to transcribe, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
Builder of TranslationRequest instances.
The audio file to transcribe, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
Builder of TranslationRequest instances.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
The audio file to translate, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
Request to translate an audio into english.
The audio file to translate, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.
Default: json
The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.
Default: 0
Create translation response.
text format depends on TranslationRequest's responseFormat. Remaining field are provided only in case of response format verbose_json.
Request to transcribe audio into the input language.
Request to translate an audio into english.
Creates a transcription request.
Creates a translation request.
The reason why OpenAI stopped generating.
A completion generated by OpenAI.
The generated chat message.
The reason why OpenAI stopped generating.
A completion generated by OpenAI.
Create ChatCompletionFunction instance.
The description of what the function does.
Builder of ChatCompletionFunction instances.
Create ChatCompletionFunction instance.
The parameters the function accepts.
The description of what the function does.
The description of what the function does.
The parameters the functions accepts, described as a JSON Schema object. See the guide for examples, and the JSON Schema reference for documentation about the format.
The parameters the functions accepts, described as a JSON Schema object. See the guide for examples, and the JSON Schema reference for documentation about the format.
To describe a function that accepts no parameters, provide Parameters.Empty`.
Builder of ChatCompletionRequest instances.
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.
Controls how the model responds to function calls. FunctionMode.None means the model does not call a function, and responds to the end-user. FunctionMode.Auto means the model can pick between an end-user or calling a function. Specifying a particular function via FunctionMode.Named forces the model to call that function. FunctionMode.None is the default when no functions are present. FunctionMode.Auto is the default if functions are present.
A list of functions the model may generate JSON inputs for.
Creates a completion for the chat message.
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.
Controls how the model responds to function calls. FunctionMode.None means the model does not call a function, and responds to the end-user. FunctionMode.Auto means the model can pick between an end-user or calling a function. Specifying a particular function via FunctionMode.Named forces the model to call that function. FunctionMode.None is the default when no functions are present. FunctionMode.Auto is the default if functions are present.
A list of functions the model may generate JSON inputs for.
The messages to generate chat completions for.
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.
What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
Builder of ChatCompletionRequest instances.
A list of functions the model may generate JSON inputs for.
The messages to generate chat completions for.
Modify the likelihood of specified tokens appearing in the completion.
Accepts a json object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
The messages to generate chat completions for.
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.
What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
We generally recommend altering this or topP but not both.
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.
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.
Controls how the model responds to function calls. FunctionMode.None means the model does not call a function, and responds to the end-user. FunctionMode.Auto means the model can pick between an end-user or calling a function. Specifying a particular function via FunctionMode.Named forces the model to call that function. FunctionMode.None is the default when no functions are present. FunctionMode.Auto is the default if functions are present.
A list of functions the model may generate JSON inputs for.
Creates a completion for the chat message.
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.
Controls how the model responds to function calls. FunctionMode.None means the model does not call a function, and responds to the end-user. FunctionMode.Auto means the model can pick between an end-user or calling a function. Specifying a particular function via FunctionMode.Named forces the model to call that function. FunctionMode.None is the default when no functions are present. FunctionMode.Auto is the default if functions are present.
A list of functions the model may generate JSON inputs for.
The messages to generate chat completions for.
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.
What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
Modify the likelihood of specified tokens appearing in the completion.
Accepts a json object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
The messages to generate chat completions for.
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.
What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
We generally recommend altering this or topP but not both.
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.
A list of generated completions
An object containing a response from the chat completion api.
The name and arguments of a function that should be called, as generated by the model.
Generated chat message.
Create ChatMessageBuilder instance.
The name and arguments of a function that should be called, as generated by the model.
Builder of ChatMessageBuilder instances.
The name and arguments of a function that should be called, as generated by the model.
The name of the author of this message. name is required if role is `ChatRole.Function, and it should be the name of the function whose response is in the content. May contain a-z, A-Z, 0-9, and underscores, with a maximum length of 64 characters.
Create ChatMessageBuilder instance.
The name of the author of this message. name is required if role is `ChatRole.Function, and it should be the name of the function whose response is in the content. May contain a-z, A-Z, 0-9, and underscores, with a maximum length of 64 characters.
The name and arguments of a function that should be called, as generated by the model.
The messages to generate chat completions for.
The name and arguments of a function that should be called, as generated by the model.
The name of the author of this message. name is required if role is `ChatRole.Function, and it should be the name of the function whose response is in the content. May contain a-z, A-Z, 0-9, and underscores, with a maximum length of 64 characters.
The name of the author of this message. name is required if role is `ChatRole.Function, and it should be the name of the function whose response is in the content. May contain a-z, A-Z, 0-9, and underscores, with a maximum length of 64 characters.
Creates a list of ChatMessage.
Creates a ChatMessage instance.
Creates a ChatMessage instance.
The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function.
Represents a function invocation with its name and serialized arguments as generated by the model.
In scenarios such as a streaming variant of the chat API, both nameOrNull and argumentsOrNull can be null.
The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function.
The name of the function to call.
The name of the function to call.
Represents the auto mode.
Represents the none mode.
Provides default function call modes.
Represents a function call mode. The value can be any string representing a specific function call mode.
Represents a named function call mode. The name indicates a specific function that the model will call.
This interface determines how the model handles function calls.
There are several modes of operation:
Default: In this mode, the model does not invoke any function None or decides itself Auto on calling a function or responding directly to the end-user. This mode becomes default if any functions are specified.
Named: In this mode, the model will call a specific function, denoted by the name attribute.
Represents a function call mode. The value can be any string representing a specific function call mode.
Represents a named function call mode. The name indicates a specific function that the model will call.
Creates a ChatMessage instance.
Creates a list of ChatCompletionFunction.
Creates a ChatMessage instance.
Represents a no params function. Equivalent to:
{"type": "object", "properties": {}}Creates a Parameters instance using a JsonObjectBuilder.
The JsonObjectBuilder to use.
Creates a Parameters instance using a JsonObjectBuilder.
Creates a Parameters instance from a JSON string.
Deserializes Parameters from JSON format.
Custom serializer for the Parameters class.
Deserializes Parameters from JSON format.
Serializes Parameters to JSON format.
Serializes Parameters to JSON format.
Represents parameters that a function accepts, described as a JSON Schema object.
Custom serializer for the Parameters class.
The function to generate chat completion function instances.
The messages to generate chat completions for.
The messages to generate chat completions for.
A completion generated by OpenAI.
A completion generated by OpenAI.
An object containing a response from the chat completion api.
Builder of ChatCompletionFunction instances.
Creates a completion for the chat message.
Creates a completion for the chat message.
Generated chat message.
The messages to generate chat completions for.
Builder of ChatMessageBuilder instances.
Creates a list of ChatMessage.
Represents a function invocation with its name and serialized arguments as generated by the model.
This interface determines how the model handles function calls.
Creates a list of ChatCompletionFunction.
Represents parameters that a function accepts, described as a JSON Schema object.
The function to generate chat completion function instances.
The messages to generate chat completions for.
The messages to generate chat completions for.
The reason the model stopped generating tokens. This will be FinishReason.Stop if the model hit a natural stop point or a provided stop sequence, or FinishReason.Length if the maximum number of tokens specified in the request was reached.
A completion generated by GPT-3.
The reason the model stopped generating tokens. This will be FinishReason.Stop if the model hit a natural stop point or a provided stop sequence, or FinishReason.Length if the maximum number of tokens specified in the request was reached.
The log probabilities of the chosen tokens and the top CompletionRequest.logprobs tokens.
The generated text. Will include the prompt if CompletionRequest.echo is true
removed from the response
The log probabilities of the chosen tokens and the top CompletionRequest.logprobs tokens.
The generated text. Will include the prompt if CompletionRequest.echo is true
Generates bestOf completions server-side and returns the "best" (the one with the lowest log probability per token). Results cannot be streamed.
When used with n, bestOf controls the number of candidate completions and n specifies how many to return, bestOf must be greater than n.
Note: Because this parameter generates many completions, it can quickly consume your token quota. Use carefully and ensure that you have reasonable settings for maxTokens and stop.
Defaults to 1
Create CompletionRequest instance.
Number between 0 and 1 (default 0) that penalizes new tokens based on their existing frequency in the text so far. Decreases the model's likelihood to repeat the same line verbatim.
Defaults to 0.
Builder of CompletionRequest instances.
Number between 0 and 1 (default 0) that penalizes new tokens based on their existing frequency in the text so far. Decreases the model's likelihood to repeat the same line verbatim.
Include the log probabilities on the logprobs most likely tokens, as well the chosen tokens. For example, if logprobs is 10, the API will return a list of the 10 most likely tokens. The API will always return the logprob of the sampled token, so there may be up to logprobs+1 elements in the response.
Number between 0 and 1 (default 0) that penalizes new tokens based on whether they appear in the text so far. Increases the model's likelihood to talk about new topics.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
Create CompletionRequest instance.
Modify the likelihood of specified tokens appearing in the completion.
Accepts a json object that maps tokens (specified by their token ID in the GPT tokenizer) to an associated bias` value from -100 to 100. You can use this tokenizer tool (which works for both GPT-2 and GPT-3) to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
As an example, you can pass {"50256": -100} to prevent the `<|endoftext|> token from being generated.
Defaults to null.
Include the log probabilities on the logprobs most likely tokens, as well the chosen tokens. For example, if logprobs is 10, the API will return a list of the 10 most likely tokens. The API will always return the logprob of the sampled token, so there may be up to logprobs+1 elements in the response.
Defaults to null.
Number between 0 and 1 (default 0) that penalizes new tokens based on whether they appear in the text so far. Increases the model's likelihood to talk about new topics.
Defaults to 0.
The prompt(s) to generate completions for, encoded as a string, a list of strings, or a list of token lists.
Note that <|endoftext|> is the document separator that the model sees during training, so if a prompt is not specified the model will generate as if from the beginning of a new document.
Defaults to <|endoftext|>.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
We generally recommend using this or topP but not both.
Defaults to 1.
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 using this or temperature but not both.
Defaults to 1.
Generates bestOf completions server-side and returns the "best" (the one with the lowest log probability per token). Results cannot be streamed.
When used with n, bestOf controls the number of candidate completions and n specifies how many to return, bestOf must be greater than n.
Note: Because this parameter generates many completions, it can quickly consume your token quota. Use carefully and ensure that you have reasonable settings for maxTokens and stop.
Defaults to 1
Number between 0 and 1 (default 0) that penalizes new tokens based on their existing frequency in the text so far. Decreases the model's likelihood to repeat the same line verbatim.
Defaults to 0.
A request for OpenAI to generate a predicted completion for a prompt. All fields are Optional.
Number between 0 and 1 (default 0) that penalizes new tokens based on their existing frequency in the text so far. Decreases the model's likelihood to repeat the same line verbatim.
Include the log probabilities on the logprobs most likely tokens, as well the chosen tokens. For example, if logprobs is 10, the API will return a list of the 10 most likely tokens. The API will always return the logprob of the sampled token, so there may be up to logprobs+1 elements in the response.
Number between 0 and 1 (default 0) that penalizes new tokens based on whether they appear in the text so far. Increases the model's likelihood to talk about new topics.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
Modify the likelihood of specified tokens appearing in the completion.
Accepts a json object that maps tokens (specified by their token ID in the GPT tokenizer) to an associated bias` value from -100 to 100. You can use this tokenizer tool (which works for both GPT-2 and GPT-3) to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
As an example, you can pass {"50256": -100} to prevent the `<|endoftext|> token from being generated.
Defaults to null.
Include the log probabilities on the logprobs most likely tokens, as well the chosen tokens. For example, if logprobs is 10, the API will return a list of the 10 most likely tokens. The API will always return the logprob of the sampled token, so there may be up to logprobs+1 elements in the response.
Defaults to null.
Number between 0 and 1 (default 0) that penalizes new tokens based on whether they appear in the text so far. Increases the model's likelihood to talk about new topics.
Defaults to 0.
The prompt(s) to generate completions for, encoded as a string, a list of strings, or a list of token lists.
Note that <|endoftext|> is the document separator that the model sees during training, so if a prompt is not specified the model will generate as if from the beginning of a new document.
Defaults to <|endoftext|>.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
We generally recommend using this or topP but not both.
Defaults to 1.
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 using this or temperature but not both.
Defaults to 1.
The character offset from the start of the returned text for each of the chosen tokens.
The log probability of each token in tokens
A map for each index in the completion result. The map contains the top CompletionRequest.logprobs tokens and their probabilities
The character offset from the start of the returned text for each of the chosen tokens.
The log probability of each token in tokens
A map for each index in the completion result. The map contains the top CompletionRequest.logprobs tokens and their probabilities
A request for OpenAI to generate a predicted completion for a prompt. All fields are Optional.
A request for OpenAI to generate a predicted completion for a prompt. All fields are Optional.
Builder of CompletionRequest instances.
Count of completion tokens.
Count of prompts tokens.
Count of total tokens.
Delete operation response.
Organization identifier.
Create EditsRequest instance.
Builder of EditsRequest instances.
The instruction that tells the model how to edit the prompt.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
Create EditsRequest instance.
The instruction that tells the model how to edit the prompt.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
We generally recommend altering this or top_p but not both.
Defaults to 1.
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.
Defaults to 1.
The instruction that tells the model how to edit the prompt.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
The instruction that tells the model how to edit the prompt.
What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
We generally recommend altering this or top_p but not both.
Defaults to 1.
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.
Defaults to 1.
A request for OpenAI to creates a new edit for the provided input, instruction, and parameters.
Builder of EditsRequest instances.
Create EmbeddingRequest instance.
Builder of EmbeddingRequest instances.
Create EmbeddingRequest instance.
Input text to get embeddings for, encoded as an array of token. Each input must not exceed 2048 tokens in length.
Unless you are embedding code, we suggest replacing newlines (\n) in your input with a single space, as we have observed inferior results when newlines are present.
Create an embedding request.
Input text to get embeddings for, encoded as an array of token. Each input must not exceed 2048 tokens in length.
Unless you are embedding code, we suggest replacing newlines (\n) in your input with a single space, as we have observed inferior results when newlines are present.
An embedding results.
An embedding result. documentation
Create an embedding request.
Create an embedding request.
Builder of EmbeddingRequest instances.
Create embeddings response.
Represents an exception thrown when an authentication error occurs while interacting with the OpenAI API.
an instance of OpenAIError containing information about the error that occurred.
the HTTP status code associated with the error.
Represents an exception thrown when an invalid request is made to the OpenAI API.
an instance of OpenAIError containing information about the error that occurred.
the HTTP status code associated with the error.
Represents an exception thrown when an error occurs while interacting with the OpenAI API.
an instance of OpenAIError containing information about the error that occurred.
the HTTP status code associated with the error.
Represents an error object returned by the OpenAI API.
error code returned by the OpenAI API.
human-readable error message describing the error that occurred.
the parameter that caused the error, if applicable.
the type of error that occurred.
OpenAI client exception
An exception thrown in case of an I/O error
Represents an exception thrown when a permission error occurs while interacting with the OpenAI API.
an instance of OpenAIError containing information about the error that occurred.
the HTTP status code associated with the error.
Represents an exception thrown when the OpenAI API rate limit is exceeded.
an instance of OpenAIError containing information about the error that occurred.
the HTTP status code associated with the error.
Represents an exception thrown when an unknown error occurs while interacting with the OpenAI API. This exception is used when the specific type of error is not covered by the existing subclasses.
an instance of OpenAIError containing information about the error that occurred.
the HTTP status code associated with the error.
Represents an exception thrown when an authentication error occurs while interacting with the OpenAI API.
An exception thrown in case of an I/O error
Represents an exception thrown when an invalid request is made to the OpenAI API.
Represents an exception thrown when an error occurs while interacting with the OpenAI API.
Represents an error response from the OpenAI API.
Represents an error object returned by the OpenAI API.
OpenAI client exception
Runtime Http Client exception
An exception thrown in case of an I/O error
An exception thrown in case of a server error
An exception thrown in case a request times out.
Represents an exception thrown when a permission error occurs while interacting with the OpenAI API.
Represents an exception thrown when the OpenAI API rate limit is exceeded.
Represents an exception thrown when an unknown error occurs while interacting with the OpenAI API. This exception is used when the specific type of error is not covered by the existing subclasses.
Creates the FileSource instance
Builder of FileSource instances.
Creates the FileSource instance
File source.
Create FileSource instance.
file path to upload
file system to be used
Represents a file resource.
Create FileSource instance.
File source.
Create a new instance of FileUpload.
The JSON Lines file to be uploaded.
If the purpose is set to "fine-tune", each line is a JSON record with "prompt" and "completion" fields representing your training examples.
Builders of FileUpload instances.
Create a new instance of FileUpload.
The intended purpose of the uploaded documents.
Use "fine-tune" for Fine-tuning. This allows us to validate the format of the uploaded file.
The JSON Lines file to be uploaded.
If the purpose is set to "fine-tune", each line is a JSON record with "prompt" and "completion" fields representing your training examples.
The intended purpose of the uploaded documents.
Use "fine-tune" for Fine-tuning. This allows us to validate the format of the uploaded file.
File status details
Represents a file resource.
Request to upload a file.
Represents a file resource.
Builder of FileSource instances.
File status.
Request to upload a file.
Builders of FileUpload instances.
The batch size to use for training. The batch size is the number of training examples used to train a single forward and backward pass.
By default, the batch size will be dynamically configured to be ~0.2% of the number of examples in the training set, capped at 256 - in general, we've found that larger batch sizes tend to work better for larger datasets.
Defaults to null.
Create a new instance of FineTuneRequest.
If this is provided, we calculate F-beta scores at the specified beta values. The F-beta score is a generalization of F-1 score. This is only used for binary classification.
With a beta of 1 (i.e. the F-1 score), precision and recall are given the same weight. A larger beta score puts more weight on recall and less on precision. A smaller beta score puts more weight on precision and less on recall.
Defaults to null.
The number of classes in a classification task.
This parameter is required for multiclass classification.
Defaults to null.
The positive class in binary classification.
This parameter is needed to generate precision, recall, and F1 metrics when doing binary classification.
Defaults to null.
If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file.
In order to compute classification metrics, you must provide a validation_file. Additionally, you must specify classification_n_classes for multiclass classification or classification_positive_class for binary classification.
Defaults to false.
Builder of FineTuneRequest instances.
If this is provided, we calculate F-beta scores at the specified beta values. The F-beta score is a generalization of F-1 score. This is only used for binary classification.
The number of classes in a classification task.
The positive class in binary classification.
If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file.
The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value.
The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short.
The ID of an uploaded file that contains training data.
The ID of an uploaded file that contains validation data.
Create a new instance of FineTuneRequest.
The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value.
By default, the learning rate multiplier is the 0.05, 0.1, or 0.2 depending on final batch_size (larger learning rates tend to perform better with larger batch sizes). We recommend experimenting with values in the range 0.02 to 0.2 to see what produces the best results.
Defaults to null.
The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short.
If prompts are extremely long (relative to completions), it may make sense to reduce this weight to avoid over-prioritizing learning the prompt.
Defaults to 0.1.
The ID of an uploaded file that contains training data.
Your dataset must be formatted as a JSONL file, where each training example is a JSON object with the keys prompt and completion. Additionally, you must upload your file with the purpose fine-tune.
The ID of an uploaded file that contains validation data.
If you provide this file, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in the fine-tuning results file. Your train and validation data should be mutually exclusive.
Your dataset must be formatted as a JSONL file, where each validation example is a JSON object with the keys prompt and completion. Additionally, you must upload your file with the purpose fine-tune.
The batch size to use for training. The batch size is the number of training examples used to train a single forward and backward pass.
By default, the batch size will be dynamically configured to be ~0.2% of the number of examples in the training set, capped at 256 - in general, we've found that larger batch sizes tend to work better for larger datasets.
Defaults to null.
If this is provided, we calculate F-beta scores at the specified beta values. The F-beta score is a generalization of F-1 score. This is only used for binary classification.
With a beta of 1 (i.e. the F-1 score), precision and recall are given the same weight. A larger beta score puts more weight on recall and less on precision. A smaller beta score puts more weight on precision and less on recall.
Defaults to null.
The number of classes in a classification task.
This parameter is required for multiclass classification.
Defaults to null.
The positive class in binary classification.
This parameter is needed to generate precision, recall, and F1 metrics when doing binary classification.
Defaults to null.
If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file.
In order to compute classification metrics, you must provide a validation_file. Additionally, you must specify classification_n_classes for multiclass classification or classification_positive_class for binary classification.
Defaults to false.
Create a Fine-Tune request.
If this is provided, we calculate F-beta scores at the specified beta values. The F-beta score is a generalization of F-1 score. This is only used for binary classification.
The number of classes in a classification task.
The positive class in binary classification.
If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file.
The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value.
The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short.
The ID of an uploaded file that contains training data.
The ID of an uploaded file that contains validation data.
The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value.
By default, the learning rate multiplier is the 0.05, 0.1, or 0.2 depending on final batch_size (larger learning rates tend to perform better with larger batch sizes). We recommend experimenting with values in the range 0.02 to 0.2 to see what produces the best results.
Defaults to null.
The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short.
If prompts are extremely long (relative to completions), it may make sense to reduce this weight to avoid over-prioritizing learning the prompt.
Defaults to 0.1.
The ID of an uploaded file that contains training data.
Your dataset must be formatted as a JSONL file, where each training example is a JSON object with the keys prompt and completion. Additionally, you must upload your file with the purpose fine-tune.
The ID of an uploaded file that contains validation data.
If you provide this file, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in the fine-tuning results file. Your train and validation data should be mutually exclusive.
Your dataset must be formatted as a JSONL file, where each validation example is a JSON object with the keys prompt and completion. Additionally, you must upload your file with the purpose fine-tune.
List of FineTuneEvents.
Fine-tuned model.
Hyper parameters.
A unique id assigned to this fine-tune.
Fine-tune of a specified model from a given dataset
List of FineTuneEvents.
Fine-tuned model.
Hyper parameters.
A unique id assigned to this fine-tune.
Organization ID.
Result Files.
List of training Files.
List of validation Files.
Organization ID.
Result Files.
List of training Files.
List of validation Files.
The number of classes in a classification task.
The positive class in binary classification. This parameter is needed to generate precision, recall, and F1 metrics when doing binary classification.
If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file.
In order to compute classification metrics, you must provide a validation_file. Additionally, you must specify classificationNClasses for multiclass classification or classificationPositiveClass for binary classification.
Fine-Tune hyper parameters.
The number of classes in a classification task.
The positive class in binary classification. This parameter is needed to generate precision, recall, and F1 metrics when doing binary classification.
If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file.
The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value.
The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short.
The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value.
The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short.
Create a Fine-Tune request.
Fine-tune of a specified model from a given dataset
Fine tune event.
Create a Fine-Tune request.
Builder of FineTuneRequest instances.
Fine-Tune hyper parameters.
Data class representing a fine-tuning job event.
The name of the fine-tuned model that is being created, or null if the fine-tuning job is still running.
The Unix timestamp (in seconds) for when the fine-tuning job was finished, or null if still running.
The hyperparameters used for the fine-tuning job.
The object identifier, which can be referenced in the API endpoints.
A data class representing a fine-tuning job.
The name of the fine-tuned model that is being created, or null if the fine-tuning job is still running.
The Unix timestamp (in seconds) for when the fine-tuning job was finished, or null if still running.
The hyperparameters used for the fine-tuning job.
The object identifier, which can be referenced in the API endpoints.
The organization that owns the fine-tuning job.
The compiled results file ID(s) for the fine-tuning job, retrievable via the Files API.
The current status of the fine-tuning job (e.g., Status.ValidatingFiles, Status.Queued, etc.).
The total number of billable tokens processed by this fine-tuning job, or null if the job is still running.
The file ID used for training, retrievable via the Files API.
The file ID used for validation, retrievable via the Files API, or null if not available.
The organization that owns the fine-tuning job.
The compiled results file ID(s) for the fine-tuning job, retrievable via the Files API.
The current status of the fine-tuning job (e.g., Status.ValidatingFiles, Status.Queued, etc.).
The total number of billable tokens processed by this fine-tuning job, or null if the job is still running.
The file ID used for training, retrievable via the Files API.
The file ID used for validation, retrievable via the Files API, or null if not available.
Create a new instance of FineTuningRequest.
The hyperparameters used for the fine-tuning job (Optional).
Builder of FineTuningRequest instances.
The hyperparameters used for the fine-tuning job (Optional).
The ID of an uploaded file that contains training data. See /docs/api-reference/files/upload for how to upload a file.
The ID of an uploaded file that contains validation data (Optional). The same data should not be present in both train and validation files.
Create a new instance of FineTuningRequest.
The ID of an uploaded file that contains training data. See /docs/api-reference/files/upload for how to upload a file.
The ID of an uploaded file that contains validation data (Optional). The same data should not be present in both train and validation files.
The hyperparameters used for the fine-tuning job.
A data class representing a fine-tuning request.
The hyperparameters used for the fine-tuning job.
The ID of an uploaded file that contains training data. See /docs/api-reference/files/upload for how to upload a file.
The ID of an uploaded file that contains validation data.
The ID of an uploaded file that contains training data. See /docs/api-reference/files/upload for how to upload a file.
Your dataset must be formatted as a JSONL file. Additionally, you must upload your file with the purpose fine-tune. See the /docs/guides/fine-tuning for more details.
The ID of an uploaded file that contains validation data.
If you provide this file, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in the fine-tuning results file. The same data should not be present in both train and validation files.
Your dataset must be formatted as a JSONL file. You must upload your file with the purpose fine-tune. See the /docs/guides/fine-tuning for more details.
A predefined NEpochs instance which indicates automatic determination of epochs.
A data class representing hyperparameters used during the fine-tuning of a model.
This class holds configuration options that guide the training process, and it supports serialization to allow for easy storage and retrieval of the settings.
The number of training epochs.
This parameter can either be a specific number or the string "auto", where "auto" implies that the number of epochs will be determined automatically. It uses a sealed interface NEpochs to accept either an integer or a string value.
Create a Fine-Tuning request.
Fine-tuning identifier.
A data class representing a fine-tuning job.
Data class representing a fine-tuning job event.
A data class representing a fine-tuning request.
Builder of FineTuningRequest instances.
A data class representing hyperparameters used during the fine-tuning of a model.
time period required to process an HTTP call: from sending a request to receiving a response
time period in which a client should establish a connection with a server
maximum time of inactivity between two data packets when exchanging data with a server
Http operations timeouts.
time period required to process an HTTP call: from sending a request to receiving a response
time period in which a client should establish a connection with a server
maximum time of inactivity between two data packets when exchanging data with a server
Image generation request. Results are expected as URLs.
Response format as base 64 json.
Response format as url.
Image generation request. Results are expected as URLs.
The format in which the generated images are returned.
Creates the ImageCreation instance
Builder of ImageCreation instances.
Creates the ImageCreation instance
Image generation request.
The image to edit. Must be a valid PNG file, less than 4MB, and square. If mask is not provided, image must have transparency, which will be used as the mask.
Builder of ImageEdit instances.
The image to edit. Must be a valid PNG file, less than 4MB, and square. If mask is not provided, image must have transparency, which will be used as the mask.
Image edit request.
Size image of dimension 1024x1024.
Creates the ImageVariation instance
The image to use as the basis for the variation(s). Must be a valid PNG file, less than 4MB, and square.
Builder of ImageVariation instances.
Creates the ImageVariation instance
The image to use as the basis for the variation(s). Must be a valid PNG file, less than 4MB, and square.
Image variant request.
Image generation request.
Image edit request.
Image variant request.
Image generation request.
Builder of ImageCreation instances.
Image edit request.
Builder of ImageEdit instances.
Image variant request.
Builder of ImageVariation instances.
Returns the enum constant of this type with the specified name. The string must match exactly an identifier used to declare an enum constant in this type. (Extraneous whitespace characters are not permitted.)
if this enum type has no constant with the specified name
Returns the enum constant of this type with the specified name. The string must match exactly an identifier used to declare an enum constant in this type. (Extraneous whitespace characters are not permitted.)
if this enum type has no constant with the specified name
Model permission details.
Model permission details.
Harassment content that also includes violence or serious harm towards any target.
Content that expresses, incites, or promotes harassing language towards any target.
Hateful content that also includes violence or serious harm towards the targeted group.
Content that expresses, incites, or promotes harassing language towards any target.
Harassment content that also includes violence or serious harm towards any target.
Hateful content that also includes violence or serious harm towards the targeted group.
Content that encourages performing acts of self-harm, such as suicide, cutting, and eating disorders, or that gives instructions or advice on how to commit such acts.
Content where the speaker expresses that they are engaging or intend to engage in acts of self-harm, such as suicide, cutting, and eating disorders.
Sexual content that includes an individual who is under 18 years old.
Violent content that depicts death, violence, or serious physical injury in extreme graphic detail.
Content that encourages performing acts of self-harm, such as suicide, cutting, and eating disorders, or that gives instructions or advice on how to commit such acts.
Content where the speaker expresses that they are engaging or intend to engage in acts of self-harm, such as suicide, cutting, and eating disorders.
Sexual content that includes an individual who is under 18 years old.
Violent content that depicts death, violence, or serious physical injury in extreme graphic detail.
The score for the category 'harassment/threatening'.
The score for the category 'harassment'.
Hateful content that also includes violence or serious harm towards the targeted group.
The score for the category 'harassment'.
The score for the category 'harassment/threatening'.
Hateful content that also includes violence or serious harm towards the targeted group.
The score for the category 'self-harm/instructions'.
The score for the category 'self-harm/intent'.
Sexual content that includes an individual who is under 18 years old.
Violent content that depicts death, violence, or serious physical injury in extreme graphic detail.
The score for the category 'self-harm/instructions'.
The score for the category 'self-harm/intent'.
Sexual content that includes an individual who is under 18 years old.
Violent content that depicts death, violence, or serious physical injury in extreme graphic detail.
Advanced notice is provided before updating this model.
Ensures you are always using the most accurate model.
Creates the ModerationRequest instance
Data class representing a ModerationRequest
Moderation model. Defaults to ModerationModel.Latest.
Creates the ModerationRequest instance
Moderation model. Defaults to ModerationModel.Latest.
Request to classify if text violates OpenAI's Content Policy.
Moderation model. Defaults to ModerationModel.Latest.
Moderation model. Defaults to ModerationModel.Latest.
Per-category binary content policy violation flags.
Per-category raw scores output by the model, denoting the model's confidence that the input violates the OpenAI's policy for the category. The value is between 0 and 1, where higher values denote higher confidence. The scores should not be interpreted as probabilities.
Per-category binary content policy violation flags.
Per-category raw scores output by the model, denoting the model's confidence that the input violates the OpenAI's policy for the category. The value is between 0 and 1, where higher values denote higher confidence. The scores should not be interpreted as probabilities.
Moderation response.
Model used for moderation response.
Moderation results.
Moderation model.
Request to classify if text violates OpenAI's Content Policy.
Data class representing a ModerationRequest
Moderation response.
Request to classify if text violates OpenAI's Content Policy.
This annotation marks a library API as beta.
Any usage of a declaration annotated with @BetaOpenAI must be accepted either by annotating that usage with the OptIn annotation, e.g. @OptIn(BetaOpenAI::class), or by using the compiler argument -Xopt-in=com.aallam.openai.api.BetaOpenAI.
This annotation marks a library API as experimental.
Any usage of a declaration annotated with @ExperimentalOpenAI must be accepted either by annotating that usage with the OptIn annotation, e.g. @OptIn(ExperimentalOpenAI::class), or by using the compiler argument -Xopt-in=com.aallam.openai.api.ExperimentalOpenAI.
API marked with this annotation is internal, and it is not intended to be used outside OpenAI API client. It could be modified or removed without any notice. Using it could cause undefined behaviour and/or any unexpected effects.
This annotation marks a library API as beta.
Any usage of a declaration annotated with @BetaOpenAI must be accepted either by annotating that usage with the OptIn annotation, e.g. @OptIn(BetaOpenAI::class), or by using the compiler argument -Xopt-in=com.aallam.openai.api.BetaOpenAI.
A marker annotations for DSLs.
This annotation marks a library API as beta.
This annotation marks a library API as experimental.
API marked with this annotation is internal, and it is not intended to be used outside OpenAI API client. It could be modified or removed without any notice. Using it could cause undefined behaviour and/or any unexpected effects.
This annotation marks a library API as beta.
A marker annotations for DSLs.