Chat
Last updated
Was this helpful?
Last updated
Was this helpful?
POST /chat/completions
Generate a chat message completion from the selected LLM.
Portkey automatically transforms the parameters for LLMs other than OpenAI. If some parameters don't exist in the other LLMs, they will be dropped.
The chat.completions.create method in the Portkey SDK enables you to generate chat completions using various Large Language Models (LLMs). This method is designed to be similar to the OpenAI chat completions API, offering a familiar interface for those accustomed to OpenAI's services.
The chat completions endpoint accepts an array of message objects and returns the completion in a chat message format.
The chat completions API also supports adding images to the request for vision models (GPT-4V, Gemini, etc).
Pass the stream parameter as true in the request to enable streaming responses from the Chat completions API.
The chat completions endpoint accepts an array of message objects and returns the completion in a chat message format.
The tools parameter accepts functions which can be sent specifically for models that support function calling.
The response will conform to the Chat Completions Object schema from the Portkey API, typically including generated messages and relevant metadata.
The body is similar to the of OpenAI and the response will be the . When choosing stream:true the response will be a stream of objects.
For REST API examples, scroll .
requestParams (Object): Parameters for the chat completion request, detailing the chat interaction. These are similar to the . Portkey automatically transforms the parameters for LLMs other than OpenAI. If some parameters don't exist in the other LLMs, they will be dropped. Portkey is multimodal by-default, so parameters relevant to vision models, like image_url, base64 data are also supported.
configParams(Object): Additional configuration options for the request. This is an optional parameter that can include custom config options for this specific request. These will override the configs set in the Portkey Client. A full list of these config parameters can be found .
To send chat requests to locally or privately hosted models, check out the guide on .
There might be a need to override values per request, or send options for and as part of the request being made. This is possible by attaching adding these parameters along with the request being made.
An ID you can pass to refer to one or more requests later on. If not provided, Portkey generates a trace ID automatically for each request. Docs
An ID you can pass to refer to a span under a trace.
Link a child span to a parent span
Name for the Span ID
Pass any arbitrary metadata along with your request
Partition your Portkey cache store based on custom strings, ignoring metadata and other headers
Forces a cache refresh for your request by making a new API call and storing the updated value
A list of messages comprising the conversation so far. Example Python code.
ID of the model to use. See the model endpoint compatibility table for details on which models work with the Chat API.
gpt-4-turboNumber 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.
See more information about frequency and presence penalties.
0Modify 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.
nullWhether 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.
falseAn integer between 0 and 20 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.
The maximum number of tokens that can be generated in the chat completion.
The total length of input tokens and generated tokens is limited by the model's context length. Example Python code for counting tokens.
How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep n as 1 to minimize costs.
1Example: 1Number 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.
See more information about frequency and presence penalties.
0An object specifying the format that the model must output.
Setting to { "type": "json_schema", "json_schema": {...} }enables Structured Outputs which ensures the model will match your
supplied JSON schema. Works across all the providers that support this functionality. OpenAI & Azure OpenAI, Gemini & Vertex AI.
Setting to { "type": "json_object" } enables the older JSON mode, which ensures the message the model generates is valid JSON.
Using json_schema is preferred for models that support it.
Default response format. Used to generate text responses.
JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
JSON object response format. An older method of generating JSON responses.
Using json_schema is recommended for models that support it. Note that the
model will not generate JSON without a system or user message instructing it
to do so.
This feature is in Beta.
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.
Up to 4 sequences where the API will stop generating further tokens.
nullIf 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. Example Python code.
falseOptions for streaming response. Only set this when you set stream: true.
nullView the thinking/reasoning tokens as part of your response. Thinking models produce a long internal chain of thought before generating a response. Supported only for specific Claude models on Anthropic, Google Vertex AI, and AWS Bedrock. Requires setting strict_openai_compliance = false in your API call.
{"type":"enabled","budget_tokens":2030}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.
1Example: 1An 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.
1Example: 1A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported.
Controls which (if any) tool is called by the model.
none means the model will not call any tool and instead generates a message.
auto means the model can pick between generating a message or calling one or more tools.
required means the model must call one or more tools.
Specifying a particular tool via {"type": "function", "function": {"name": "my_function"}} forces the model to call that tool.
none is the default when no tools are present. auto is the default if tools are present.
none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools.
Specifies a tool the model should use. Use to force the model to call a specific function.
Whether to enable parallel function calling during tool use.
trueA unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. Learn more.
user-1234Deprecated in favor of tool_choice.
Controls which (if any) function is called by the model.
none means the model will not call a function and instead generates a message.
auto means the model can pick between generating a message or calling a function.
Specifying a particular function via {"name": "my_function"} forces the model to call that function.
none is the default when no functions are present. auto is the default if functions are present.
none means the model will not call a function and instead generates a message. auto means the model can pick between generating a message or calling a function.
Specifying a particular function via {"name": "my_function"} forces the model to call that function.
Deprecated in favor of tools.
A list of functions the model may generate JSON inputs for.
Represents a chat completion response returned by model, based on the provided input.
OK