model inference settings

packages/orchestration-service/src/routes/models.js

@@ -4,6 +4,9 @@ const fs = require('fs');
 const path = require('path');
 const appSettings = require('../config/settings');
 
+const { getEnv, SERVICES } = require('@nexusai/shared');
+const INFERENCE_SERVICE_URL = getEnv('INFERENCE_SERVICE_URL', SERVICES.INFERENCE_URL);
+
 router.get('/', (req, res) => {
   const { modelsFolderPath } = appSettings.load();
 
@@ -40,6 +43,21 @@ router.get('/', (req, res) => {
   }
 });
 
+router.get('/props', async (req, res) => {
+  try {
+    const response = await fetch(`${INFERENCE_SERVICE_URL}/props`);
+    if (!response.ok) throw new Error(`Inference service error: ${response.status}`);
+    const data = await response.json();
+    res.json({
+      contextWindow: data.n_ctx,
+      modelAlias: data.model_alias,
+    });
+  } catch (err) {
+    console.error('[models/props]', err.message);
+    res.status(503).json({ error: 'Could not reach inference service' });
+  }
+});
+
 function getFileSizeMB(filepath) {
   try {
     const bytes = fs.statSync(filepath).size;