Azure OpenAI (Preview)

Deployment ID of the deployed model

API version

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 top_p 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.

If set, partial message deltas will be sent, like in ChatGPT. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a data: [DONE] message.

Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.

The maximum number of tokens allowed for the generated answer. By default, the number of tokens the model can return will be (4096 - prompt tokens).

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.

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.

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.

A unique identifier representing your end-user, which can help Azure OpenAI to monitor and detect abuse.

The role of the messages author.

An array of content parts with a defined type, each can be of type text.

A representation of configuration data for a single Azure OpenAI chat extension. This will be used by a chat completions request that should use Azure OpenAI chat extensions to augment the response behavior. The use of this configuration is compatible only with Azure OpenAI.

The configured top number of documents to feature for the configured query.

Whether queries should be restricted to use of indexed data.

The configured strictness of the search relevance filtering. The higher of strictness, the higher of the precision but lower recall of the answer.

Give the model instructions about how it should behave and any context it should reference when generating a response. You can describe the assistant's personality and tell it how to format responses. There's a 100 token limit for it, and it counts against the overall token limit.

The name of the index to use as available in the referenced Azure Search resource.

The name of the index field to use as a title.

The name of the index field to use as a URL.

The name of the index field to use as a filepath.

The names of index fields that should be treated as content.

The separator pattern that content fields should use.

The names of fields that represent vector data.

The type of Azure Search retrieval query that should be executed when using it as an Azure OpenAI chat extension.

The additional semantic configuration for the query.

Search filter.

Represents the available sources Azure OpenAI On Your Data can use to configure vectorization of data for use with vector search.

Specifies the name of the model deployment to use for vectorization. This model deployment must be in the same Azure OpenAI resource, but On Your Data will use this model deployment via an internal call rather than a public one, which enables vector search even in private networks.

How many chat completion choices to generate for each input message.

If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result.Determinism is not guaranteed, and you should refer to the system_fingerprint response parameter to monitor changes in the backend.

Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message. This option is currently not available on the gpt-4-vision-preview model.

An integer between 0 and 5 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to true if this parameter is used.

Setting to json_object enables JSON mode. This guarantees that the message the model generates is valid JSON.

Deployment ID of the deployed model

API version

Input text to get embeddings for, encoded as a string. Input string must not exceed 2048 tokens in length

A unique identifier representing your end-user, which can help Azure OpenAI to monitor and detect abuse.

The input type of embedding search to use.

Deployment ID of the deployed model

Confirm Deployment ID of the deployed model

API version

The index of the message in the conversation.

The role of the author of this message.

The recipient of the message in the format of .. Present if and only if the recipient is tool.

The contents of the message

Whether the message ends the turn.

The data source type.

The parameters to be used for the data source in runtime.

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 top_p 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.

If set, partial message deltas will be sent, like in ChatGPT. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a data: [DONE] message.

Array minimum size of 1 and maximum of 4

The maximum number of tokens allowed for the generated answer. By default, the number of tokens the model can return will be (4096 - prompt tokens).

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.

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.

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.

A unique identifier representing your end-user, which can help Azure OpenAI to monitor and detect abuse.