Maxim

ConfigDict

class ConfigDict(TypedDict)

A class representing the configuration for the Maxim SDK. Attributes:

Name	Type	Description
`api_key`	Optional[str], optional	The API key for the Maxim instance. Defaults to None.
`base_url`	Optional[str], optional	The base URL for the Maxim instance. Defaults to None.
`cache`	Optional[MaximCache], optional	The cache to use for the Maxim instance. Defaults to None.
`debug`	Optional[bool], optional	Whether to enable debug logging. Defaults to False.
`raise_exceptions`	Optional[bool], optional	Whether to raise exceptions during logging operations. Defaults to False.
`prompt_management`	Optional[bool], optional	Whether to enable prompt management. Defaults to False.

Config

@deprecated(
    "This class will be removed in a future version. Use {} which is TypedDict."
)

@dataclass
class Config()

A class representing the configuration for the Maxim SDK. Attributes:

Name	Type	Description
`api_key`	str	The API key for the Maxim instance.
`base_url`	Optional[str], optional	The base URL for the Maxim instance. Defaults to “https://app.getmaxim.ai”.
`cache`	Optional[MaximCache], optional	The cache to use for the Maxim instance. Defaults to None.
`debug`	Optional[bool], optional	Whether to enable debug logging. Defaults to False.
`raise_exceptions`	Optional[bool], optional	Whether to raise exceptions during logging operations. Defaults to False.
`prompt_management`	Optional[bool], optional	Whether to enable prompt management. Defaults to False.

get_config_dict

def get_config_dict(config: Union[Config, ConfigDict]) -> dict[str, Any]

Converts a Config or ConfigDict to a dictionary with default values. Arguments:

Name	Type	Description
`config`	Union[Config, ConfigDict]	The configuration object to convert.

Returns: dict[str, Any]: A dictionary containing the configuration parameters with defaults applied.

class Maxim()

init

def __init__(config: Union[Config, ConfigDict, None] = None)

Initializes a new instance of the Maxim class. Arguments:

Name	Type	Description
`config`	Config	The configuration for the Maxim instance.

enable_prompt_management

def enable_prompt_management(cache: Optional[MaximCache] = None) -> "Maxim"

Enables prompt management functionality with optional cache configuration. Arguments:

Name	Type	Description
`cache`	Optional[MaximCache], optional	Custom cache implementation to use.

Defaults to None (uses existing cache). Returns:

Name	Description
`[Maxim](/sdk/python/references/maxim)`	The current Maxim instance for method chaining.

enable_exceptions

def enable_exceptions(val: bool) -> "Maxim"

Enables or disables exception raising during logging operations. Arguments:

Name	Type	Description
`val`	bool	True to enable exception raising, False to disable.

Returns:

Name	Description
`[Maxim](/sdk/python/references/maxim)`	The current Maxim instance for method chaining.

get_prompt

def get_prompt(id: str, rule: QueryRule) -> Optional[RunnablePrompt]

Retrieves a prompt based on the provided id and rule. Arguments:

Name	Type	Description
`id`	str	The id of the prompt.
`rule`	QueryRule	The rule to match the prompt against.

Returns:

Name	Description
`Optional[[Prompt](/sdk/python/references/models/prompt)]`	The prompt object if found, otherwise None.

get_prompts

def get_prompts(rule: QueryRule) -> List[RunnablePrompt]

Retrieves all prompts that match the given rule. Arguments:

Name	Type	Description
`rule`	QueryRule	The rule to match the prompts against.

Returns:

Name	Description
`List[[Prompt](/sdk/python/references/models/prompt)]`	A list of prompts that match the given rule.

get_prompt_chain

def get_prompt_chain(id: str,
                     rule: QueryRule) -> Optional[RunnablePromptChain]

Retrieves a prompt chain based on the provided id and rule. Arguments:

Name	Type	Description
`id`	str	The id of the prompt chain.
`rule`	QueryRule	The rule to match the prompt chain against.

Returns:

Name	Description
`Optional[[PromptChain](/sdk/python/references/models/prompt_chain)]`	The prompt chain object if found, otherwise None.

get_folder_by_id

def get_folder_by_id(id: str) -> Optional[Folder]

Retrieves a folder based on the provided id. Arguments:

Name	Type	Description
`id`	str	The id of the folder.

Returns:

Name	Description
`Optional[[Folder](/sdk/python/references/models/folder)]`	The folder object if found, otherwise None.

get_folders

def get_folders(rule: QueryRule) -> List[Folder]

Retrieves all folders that match the given rule. Arguments:

Name	Type	Description
`rule`	QueryRule	The rule to match the folders against.

Returns:

Name	Description
`List[[Folder](/sdk/python/references/models/folder)]`	A list of folders that match the given rule.

logger

def logger(
        config: Optional[Union[LoggerConfig,
                               LoggerConfigDict]] = None) -> Logger

Creates a logger based on the provided configuration. Arguments:

Name	Type	Description
`config`	LoggerConfig	The configuration for the logger.

Returns:

Name	Description
`[Logger](/sdk/python/references/logger/logger)`	The logger object.

create_test_run

def create_test_run(name: str, in_workspace_id: str) -> TestRunBuilder

Creates a test run builder based on the provided name and workspace id. Arguments:

Name	Type	Description
`name`	str	The name of the test run.
`in_workspace_id`	str	The workspace id to create the test run in.

Returns:

Name	Description
`[TestRunBuilder](/sdk/python/references/test_runs/test_run_builder)`	The test run builder object.

chat_completion

def chat_completion(model: str,
                    messages: List[ChatCompletionMessage],
                    tools: Optional[List[Tool]] = None,
                    **kwargs) -> Optional[PromptResponse]

Performs a chat completion request using the specified model and messages. Arguments:

Name	Type	Description
`model`	str	The model name to use for completion. The expected format is “provider/model_name”. Example “openai/gpt-3.5-turbo”.
`messages`	List[ChatCompletionMessage]	List of chat messages in the conversation
`tools`	Optional[List[Tool]], optional	List of tools available to the model. Defaults to None.
`**kwargs`		Additional model parameters to pass to the completion request

Returns:

Name	Description
`Optional[[PromptResponse](/sdk/python/references/models/prompt)]`	The completion response if successful, None otherwise

cleanup

def cleanup()

Cleans up the Maxim sync thread.

Overview

Python

Typescript

Maxim

ConfigDict

Config

get_config_dict

Maxim

init

enable_prompt_management

enable_exceptions

get_prompt

get_prompts

get_prompt_chain

get_folder_by_id

get_folders

logger

create_test_run

chat_completion

cleanup

Overview

Python

Typescript

​ConfigDict

​Config

​get_config_dict

​Maxim

​__init__

​enable_prompt_management

​enable_exceptions

​get_prompt

​get_prompts

​get_prompt_chain

​get_folder_by_id

​get_folders

​logger

​create_test_run

​chat_completion

​cleanup

ConfigDict

Config

get_config_dict

Maxim

init

enable_prompt_management

enable_exceptions

get_prompt

get_prompts

get_prompt_chain

get_folder_by_id

get_folders

logger

create_test_run

chat_completion

cleanup