OPEA API Service Spec (v0.9)¶
Authors:
This specification is used to define the Restful API of OPEA Mega Service for users to access, as long as the input and output definition of all OPEA Micro Services for developer to build OPEA Mega service.
Note
This API Service Specification is a work-in-progress and may be incomplete and contain errors.
OPEA Mega Service API¶
OPEA Mega Service is the main entry user can access for a prebuilt GenAI application. Such GenAI application consists of single or several OPEA Micro Services chained as a DAG (Directed Acyclic Graph) and built as an execution workflow for developer to create complex applications.
List Services¶
List all supported services by the OPEA Mega Service.
Request¶
Method |
URL |
---|---|
GET |
|
Response¶
Status |
Response |
---|---|
200 |
{
<service_name>: <service_description>
}
|
405 |
|
List Configurable Parameters¶
List all configurable parameters for users to control the behavior of the OPEA Mega Service.
Request¶
Method |
URL |
---|---|
GET |
|
Response¶
Status |
Response |
---|---|
200 |
{
<micro_service_name>:
{
<parameter_name>: data_type,
. . .
}
}
For example: |
405 |
|
Embedding¶
Optional. Only exists if a single OPEA microservice which exposes
/micro_service/embedding
interface is built as OPEA Mega service.
Request¶
Method |
URL |
---|---|
POST |
|
Type |
Parameters |
Values |
Required |
Description |
---|---|---|---|---|
POST |
|
|
required |
Input text to embed, encoded as a string or array of tokens. To embed multiple inputs in a single request, pass an array of strings or array of token arrays. The input must not exceed the max input tokens for the model (8192 tokens for text-embedding-ada-002), cannot be an empty string, and any array must be 2048 dimensions or less. |
POST |
|
|
deprecated |
The ID of the model to use. |
POST |
|
|
required |
The format to return the embeddings in. Can be either |
POST |
|
|
optional |
The number of dimensions the resulting output embeddings should have. Only supported in text-embedding-3 and later models. |
Response¶
Status |
Response |
---|---|
200 |
{
"object": "list",
"data": [{
"object": "embedding",
"embedding": [
0.0023064255,
...
],
"index": 0
}],
"model": "text-embedding-ada-002",
"usage": {
"prompt_tokens": 8,
"total_tokens": 8
},
}
For example: |
405 |
|
Chat¶
Optional. . If a OPEA Mega service is built with this request url, it complies with below format.
Request¶
Method |
URL |
---|---|
POST |
|
Type |
Parameters |
Values |
Required |
Description |
---|---|---|---|---|
POST |
|
|
required |
A list of messages comprising the conversation. Refer to the detail format. |
POST |
|
|
deprecated |
The ID of the model to use. |
POST |
|
|
optional |
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. |
POST |
|
|
optional |
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. |
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
Response¶
Status |
Response |
---|---|
200 |
{
"id": "chatcmpl-123",
"object": "chat.completion",
"created": 1677652288,
"model": "gpt-3.5-turbo-0125",
"system_fingerprint": "fp_44709d6fcb",
"choices": [{
"index": 0,
"object": "embedding",
"message": {
"role": "assistant",
"content": "\n\nHello there, how may I assist you today?",
},
"logprobs": null,
"finish_reason": "stop",
}],
"usage": {
"prompt_tokens": 9,
"completion_tokens": 12,
"total_tokens": 21
},
}
|
Other Operations¶
Check the usage description returned in List Services to know what other operations are supported by this OPEA Mega Service.
OPEA Micro Service API¶
OPEA Micro Service is the building block of constructing any GenAI applications. The API in OPEA micro service is used by developers to construct OPEA Mega Service like a DAG chain and is invisible for end user.
Embedding Micro Service¶
The micro service is used to generate a vector representation of a given input.
Request¶
Method |
URL |
---|---|
POST |
|
Type |
Parameters |
Values |
Required |
Description |
---|---|---|---|---|
POST |
|
|
required |
Input text to embed, encoded as a string or array of tokens. To embed multiple inputs in a single request, pass an array of strings or array of token arrays. The input must not exceed the max input tokens for the model (8192 tokens for text-embedding-ada-002), cannot be an empty string, and any array must be 2048 dimensions or less |
POST |
|
|
required |
The ID of the model to use. |
POST |
|
|
optional |
The format to return the embeddings in. Can be either |
POST |
|
|
optional |
The number of dimensions the resulting output embeddings should have. |
POST |
|
|
optional |
A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. |
Response¶
Status |
Response |
---|---|
200 |
{
"object": "list",
"data": [{
"object": "embedding",
"embedding": [
0.0023064255,
-0.009327292,
. . . (1536 floats total for ada-002)
-0.0028842222,
],
"index": 0
}],
"model": "text-embedding-ada-002",
"usage": {
"prompt_tokens": 8,
"total_tokens": 8
},
}
|
405 |
|
LLM Generation Micro Service¶
The micro service is used to provide LLM generation service.
Request¶
Method |
URL |
---|---|
POST |
|
Type |
Parameters |
Values |
Required |
Description |
---|---|---|---|---|
POST |
|
|
required |
A list of messages comprising the conversation so far. Example Python code. |
POST |
|
|
required |
The ID of the model to use. See the model endpoint compatibility table for details on which models work with the Chat API. |
POST |
|
|
optional |
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. |
POST |
|
|
optional |
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. |
POST |
|
|
optional |
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. |
POST |
|
|
optional |
An integer between 0 and 20 specifying the number of most likely tokens
to return at each token position, each with an associated log
probability. |
POST |
|
|
optional |
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. |
POST |
|
|
optional |
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. |
POST |
|
|
optional |
|
POST |
|
|
optional |
|
POST |
|
|
optional |
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 |
POST |
|
|
optional |
Specifies the latency tier to use for processing the request. This
parameter is relevant for customers subscribed to the scale tier
service. If set to |
POST |
|
|
optional |
Up to 4 sequences where the API will stop generating further tokens. |
POST |
|
|
optional |
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: |
POST |
|
|
optional |
Options for streaming response. Only set this when you set |
POST |
|
|
optional |
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
|
POST |
|
|
optional |
An alternative to sampling with temperature, called nucleus sampling,
where the model considers the results of the tokens with |
POST |
|
|
optional |
A 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. |
POST |
|
|
optional |
Controls which (if any) tool is called by the model. |
Response¶
Status |
Response |
---|---|
200 |
{
"id": "chatcmpl-123",
"object": "chat.completion",
"created": 1677652288,
"model": "gpt-4o-mini",
"system_fingerprint": "fp_44709d6fcb",
"choices": [{
"index": 0,
"object": "embedding",
"message": {
"role": "assistant",
"content": "\n\nHello there, how may I assist you today?",
},
"logprobs": null,
"finish_reason": "stop",
}],
"usage": {
"prompt_tokens": 9,
"completion_tokens": 12,
"total_tokens": 21
},
}
|
405 |
|
ASR Micro Service¶
The micro service is used to provide audio to text service.
Request¶
Method |
URL |
---|---|
POST |
|
Type |
Parameters |
Values |
Required |
Description |
---|---|---|---|---|
POST |
|
|
optional |
The link to the audio. |
POST |
|
|
optional |
The model used to do audio-to-text translation. |
POST |
|
|
optional |
The language that model prefer to detect. Default is |
Response¶
Status |
Response |
---|---|
200 |
{
"text": string
}
|
405 |
|
RAG Retrieval Micro Service¶
The micro service is used to provide RAG retrieval service. It’s usually after embedding micro sevice and before RAG reranking micro service to build a RAG Mega service.
Request¶
Method |
URL |
---|---|
POST |
|
Type |
Parameters |
Values |
Required |
Description |
---|---|---|---|---|
POST |
|
|
required |
The input string to query. |
POST |
|
|
required |
The list of float for text as vector representation. |
Response¶
Status |
Response |
---|---|
200 |
{
"retrieved_docs": list of string,
"initial_query": string,
"json_encoders": [{
"text": "I am the agent of chatbot. What can I do for you?",
},
...
]
}
|
405 |
|
RAG Reranking Micro Service¶
The micro service is used to provide RAG reranking service. It’s usually after RAG retrieval and before LLM generation micro service.
Request¶
Method |
URL |
---|---|
POST |
|
Type |
Parameters |
Values |
Required |
Description |
---|---|---|---|---|
POST |
|
|
required |
The docs to be retreived. |
POST |
|
|
required |
The string to query. |
POST |
|
|
required |
The json encoder used. |
Response¶
Status |
Response |
---|---|
200 |
{
"query": string,
"doc": [{
"text": "I am the agent of chatbot. What can I do for you?",
},
...
]
}
|
405 |
|