minor changes

src/llm_integration/content_llm.rs

@@ -1,10 +1,13 @@
 use serde::{Deserialize, Serialize};
 
+use crate::editors::context_editor::ProjectContext;
+
 pub fn continue_content(
     context: &str,
     previous_content: &str,
     instruction: &str,
-    _max_tokens: usize,
+    max_tokens: usize,
+    project: &ProjectContext,
 ) -> Result<String, Box<dyn std::error::Error>> {
     let client = reqwest::blocking::Client::new();
 
@@ -35,10 +38,12 @@ pub fn continue_content(
     let request = ChatRequest {
         messages,
         temperature: 0.7,
+        max_tokens,
+        stream: false,
     };
 
     let response = client
-        .post("http://localhost:1234/v1/chat/completions")
+        .post(project.llm_api_uri.clone() + "/v1/chat/completions")
         .json(&request)
         .send()?;
 
@@ -54,11 +59,14 @@ pub fn continue_content(
         Err("No response from model".into())
     }
 }
+
 // Simple request structure
 #[derive(Serialize)]
 struct ChatRequest {
     messages: Vec<Message>,
     temperature: f32,
+    max_tokens: usize,
+    stream: bool,
 }
 
 #[derive(Serialize, Deserialize, Debug)]