proxy setup

app.py

@@ -1,9 +1,11 @@
-from flask import Flask, render_template
+from flask import Flask, render_template, Response, stream_with_context, request
 from flask_flatpages import FlatPages
+from werkzeug.middleware.proxy_fix import ProxyFix
+import requests
 
 app = Flask(__name__)
+app.wsgi_app = ProxyFix(app.wsgi_app, x_proto=1, x_host=1)
 
-# Consolidate Configs
 app.config.update(
     FLATPAGES_AUTO_RELOAD = True,
     FLATPAGES_EXTENSION = '.md',
@@ -22,6 +24,20 @@ def index():
 def about():
     return render_template('about.html')
 
+@app.route('/proxy-chat', methods=['POST'])
+def proxy_chat():
+    target_url = "http://192.168.0.37:5002/v1/chat/completions"
+    
+    # Forward the request to your local LLM
+    # We use stream=True here to get the chunks from the backend
+    req = requests.post(target_url, json=request.json, stream=True)
+
+    def generate():
+        # Yield each chunk as it arrives from the LLM
+        for chunk in req.iter_content(chunk_size=1024):
+            yield chunk
+
+    return Response(stream_with_context(generate()), content_type=req.headers['content-type'])
 
 @app.route('/post/<path:path>/') # Adding /post/ prefix helps organize URLs
 def post(path):

requirements.txt

@@ -1,3 +1,4 @@
 flask
 flask-flatpages
 gunicorn
+requests