WecoAI Client
WecoAI ¶
A client for the WecoAI function builder API that allows users to build and query specialized functions for GenAI tasks and features. The user must simply provide a task description to build a function, and then query the function with an input to get the result they need. Our client supports both synchronous and asynchronous request paradigms and uses HTTP/2 for faster communication with the API. Support for structured outputs, multimodality, grounding through web search and reasoning is included. For maximum control/flexibility, we recommend that users build functions on the WecoAI platform to leverage the full power of the API, then deploy it in production using this client.
__init__ ¶
Initializes the WecoAI client.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
api_key | str | The API key used for authentication. If not provided, the client will attempt to read it from the environment variable - | None |
timeout | float | The timeout for the HTTP requests in seconds. Default is 120. | 120.0 |
http2 | bool | Whether to use HTTP/2 protocol for the HTTP requests. Default is True. | True |
Raises:
| Type | Description |
|---|---|
ValueError | If the API key is not provided to the client, is not set as an environment variable ( |
abuild async ¶
Build a specialized function for a task.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
task_description | str | The description of the task for which the function is being built. | required |
Returns:
| Type | Description |
|---|---|
tuple[str, int, str] | A tuple containing the name, version number and description of the function. |
aquery async ¶
Queries a specific function with the input (text, images or both) asynchronously.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
fn_name | str | The name of the function to query. | required |
version | Optional[Union[str, int]] | The version alias/number of the function to query. Default is -1 which results in the latest version being used. | -1 |
text_input | str | The text input to the function. | '' |
images_input | List[str] | A list of image URLs or base64 encoded images to be used as input to | [] |
return_reasoning | bool | A flag to indicate if the reasoning should be returned. Default is False. | False |
strict | bool | A flag to indicate if the function should be queried in strict mode where the inputs provided should match the input modalities of the LLM chosen for this function. For example, when strict is True, a text-image query to a function that uses a text-only LLM will raise an error. When strict is False, the function will attempt to handle the input by dropping the image components. Default is False. | False |
Returns:
| Type | Description |
|---|---|
dict | A dictionary containing the output of the function, the number of input tokens, the number of output tokens, and the latency in milliseconds. |
batch_query ¶
Performs batch queries on a specified function version.
This method processes multiple inputs concurrently using asynchronous queries, ensuring that results are returned in the same order as the inputs.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
fn_name | str | The name of the function to query. | required |
batch_inputs | List[Dict[str, Any]] | A list of input dictionaries for the function. Each dictionary can include: - "text_input": A string for text input. - "images_input": A list of image URLs or base64 encoded images. | required |
version | Union[str, int] | The version alias or number of the function to query. Defaults to -1 for the latest version. | -1 |
return_reasoning | bool | If True, includes reasoning in the response. Defaults to False. | False |
strict | bool | A flag to indicate if the function should be queried in strict mode where the inputs provided should match the input modalities of the LLM chosen for this function. For example, when strict is True, a text-image query to a function that uses a text-only LLM will raise an error. When strict is False, the function will attempt to handle the input by dropping the image components. Default is False. | False |
Returns:
| Type | Description |
|---|---|
List[Dict[str, Any]] | A list of dictionaries, each containing the result of a function query, including: - The function's output. - The number of input tokens. - The number of output tokens. - The latency in milliseconds. |
build ¶
Build a specialized function for a task.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
task_description | str | The description of the task for which the function is being built. | required |
Returns:
| Type | Description |
|---|---|
tuple[str, int, str] | A tuple containing the name, version number and description of the function. |
query ¶
Queries a specific function with the input (text, images or both).
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
fn_name | str | The name of the function to query. | required |
version | str | int | The version alias/number of the function to query. Default is -1 which results in the latest version being used. | -1 |
text_input | str | The text input to the function. | '' |
images_input | List[str] | A list of image URLs or base64 encoded images to be used as input to the function. | [] |
return_reasoning | bool | A flag to indicate if the reasoning should be returned. Default is False. | False |
strict | bool | A flag to indicate if the function should be queried in strict mode where the inputs provided should match the input modalities of the LLM chosen for this function. For example, when strict is True, a text-image query to a function that uses a text-only LLM will raise an error. When strict is False, the function will attempt to handle the input by dropping the image components. Default is False. | False |
Returns:
| Type | Description |
|---|---|
dict | A dictionary containing the output of the function, the number of input tokens, the number of output tokens, and the latency in milliseconds. |