Invoke large models

Command Description

Call large models to answer questions or process natural language

Command Prototype

LLM.GeneralChat({"name":"","model":"","base_url":"","api_key":""},"","","",30,0.7,4000,"text",0)

Command parameters

parameter	mandatory	type	default value	Instructions
config	True	object	{"name":"","model":"","base_url":"","api_key":""}	Configure the model by filling in the model name, model access address, and API key
content	true	string	""	user prompt
prompt	false	string	""	system prompt
file_path	false	string	""	Image path (JPEG/PNG/BMP, maximum 10MB, some models may not support all three formats of images)
timeout	false	int	30s	Timeout (seconds)
temperature	false	int	zero point seven	temperature
max_output_length	false	int	four thousand	Maximum output length (token)
format	false	object	Plain text, JSON format	Choose the format of the content returned by the large model
max_retry_count	false	int	three	When the large model returns an error or returns a structure that does not conform to the output format, the context content will be resent to the large model, allowing it to output again. The maximum number of retries is 3

Running instance

/*************************** Input Text **********************************
Command Prototype:     
    LLM.GeneralChat({"name":"","model":"","base_url":"","api_key":""},"","","",30,0.7,4000,"text",0)
    
Input Parameters:
    config -- Model configuration, specify model name, access URL, API key
    content -- User prompt
    prompt -- System prompt
    file_path -- Image path (JPEG/PNG/BMP, max 10MB, some models may not support all three formats)
    timeout -- Timeout in seconds
    temperature -- Temperature
    max_output_length -- Maximum output length (tokens)
    format -- Format of the LLM response content
    max_retry_count -- Maximum retry attempts on failure

Output Parameters:
    sRet -- Command execution result assigned to this variable

Important Notes:
    Deleting the LLM environment configuration file or removing the invoked LLM in the configuration management interface will prevent the command from running normally

*********************************************************************/

 //Using Pre-configured Model:
 LLM.GeneralChat({"name":"deepseek"},"Help me translate the content in the image into Chinese","You are a translator with professional translation expertise",'''C:\Users\Administrator\Downloads\1280X1280.PNG''',30,0.7,4000,"text",0)

 //Using Custom Model Configuration:
 LLM.GeneralChat({"model":"openai/deepseek-chat","api_key":"123456789","base_url":"https://api.deepseek.com"},"Help me translate the content in the image into Chinese","You are a translator with professional translation expertise",'''C:\Users\Administrator\Downloads\1280X1280.PNG''',30,0.7,4000,"text",0)

Visual Example