LLM request
The llm_request function emulates an LLM (Large Language Model) request. (Available from version 25.3 and later.)
C Language
int llm_request(const char *RequestName, const char *URL=<myURL>, const char *AIModel=<my_ai_model>, LAST );
Java Language
int object.llm.request(String RequestName, String URL=<myURL>, String AIModel=<my_ai_model>);
Argument | Description |
---|---|
object | An expression evaluating to an object of type WebApi. Usually web for Java. See also
Function and Constant Prefixes. |
RequestName | The title of the request. |
URL | The URL of the API, for example https://api.openai.com/v1/chat/completions |
AIModel | The AI model, for example, openai |
Return Values
This function returns LR_PASS (0) on success, and LR_FAIL (1) on failure.
Parameterization
The following argument(s) can be parameterized using standard parameterization.
Examples
We recommend that you add headers before the LLM requests. Use lr_eval_string to retrieve the body.
The following example illustrates a script with an LLM request for OpenAI.
web_add_header("Content-Type","application/json");
web_add_header("Authorization","<API_KEY>");
llm_request("OpenAI_Chat_Completion",
"URL=https://api.openai.com/v1/chat/completions",
"AIModel=openai",
lr_eval_string("{request_body}"),
LAST);
return 0;
The following example illustrates a script with an LLM request for Gemini.
lr_param_sprintf ("request_body", "Body={ \"contents\":[ { \"parts\":[{\"text\": \"hello?\"}]} ] }");
web_add_header("Content-Type","application/json");
llm_request("Gemini_Chat_Completion",
"URL=https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key=<API_KEY> ",
"AIModel=gemini",
lr_eval_string("{request_body}"),
LAST);