Update backend/main.py

backend/main.py

@@ -55,7 +55,6 @@ from rag.vector_store import VectorStore
 load_dotenv()
 
 CEREBRAS_API_KEY = os.getenv("CEREBRAS_API_KEY")
-GROQ_API_KEY = os.getenv("GROQ_API_KEY")
 YOUTUBE_API_KEY = os.getenv("YOUTUBE_API_KEY")
 
 JWT_SECRET_KEY = os.getenv(
@@ -111,22 +110,7 @@ else:
     print("⚠️ CEREBRAS_API_KEY not set — Cerebras disabled")
 
 
-# GROQ 
-groq_client = None
-
-if GROQ_API_KEY:
-    try:
-        from groq import Groq
-        groq_client = Groq(api_key=GROQ_API_KEY)
-        print("✅ Groq client initialized (secondary)")
-    except Exception as e:
-        print(f"⚠️ Groq init failed: {e}")
-        groq_client = None
-else:
-    print("ℹ️ GROQ_API_KEY not set — Groq fallback disabled")
-
-
-# OLLAMA (LAST RESORT)
+# OLLAMA (FALLBACK)
 OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://localhost:11434/v1")
 OLLAMA_MODEL = os.getenv("OLLAMA_MODEL", "llama3")
 
@@ -373,13 +357,6 @@ class UserLogin(BaseModel):
     password: str
 
 
-GROQ_MODELS = {
-    "fast": "openai/gpt-oss-20b",
-    "balanced": "openai/gpt-oss-120b",
-}
-GROQ_PREFERRED = GROQ_MODELS["balanced"]
-
-
 SYSTEM_PROMPT = """
 You are Cortexa, a powerful AI assistant.
 
@@ -411,7 +388,7 @@ IMPORTANT RULES:
 11. For coding questions, provide correct,
     practical code and explain the important parts.
 """
-# AI RESPONSE — FALLBACK CHAIN
+# AI RESPONSE — FALLBACK CHAIN (Cerebras -> Ollama)
 async def get_ai_response(message: str, context: Optional[str] = None) -> str:
     messages = [{"role": "system", "content": SYSTEM_PROMPT}]
 
@@ -439,24 +416,7 @@ async def get_ai_response(message: str, context: Optional[str] = None) -> str:
         except Exception as e:
             print(f"⚠️ Cerebras failed: {str(e)[:200]}")
 
-    # 2️⃣ Groq
-    if groq_client:
-        try:
-            completion = groq_client.chat.completions.create(
-                model=GROQ_PREFERRED,
-                messages=messages,
-                temperature=0.7,
-                max_tokens=1024,
-                stream=False,
-            )
-            text = completion.choices[0].message.content
-            if text:
-                print("✅ Response from Groq")
-                return text
-        except Exception as e:
-            print(f"⚠️ Groq failed: {str(e)[:200]}")
-
-    # 3️⃣ Ollama
+    # 2️⃣ Ollama
     try:
         async with httpx.AsyncClient(timeout=120.0) as hc:
             r = await hc.post(
@@ -939,19 +899,28 @@ Description:
             except Exception as e:
                 print(f"⚠️ Cerebras summary failed: {e}")
 
-        if groq_client:
-            completion = groq_client.chat.completions.create(
-                model=GROQ_PREFERRED,
-                messages=messages,
-                temperature=0.7,
-                max_tokens=500,
-                stream=False
-            )
-            return {
-                "summary": completion.choices[0].message.content,
-                "title": info["title"],
-                "video_id": video_id
-            }
+        # Ollama fallback for summary
+        try:
+            async with httpx.AsyncClient(timeout=120.0) as hc:
+                r = await hc.post(
+                    f"{OLLAMA_BASE_URL}/chat/completions",
+                    json={
+                        "model": OLLAMA_MODEL,
+                        "messages": messages,
+                        "temperature": 0.7,
+                    },
+                )
+                r.raise_for_status()
+                rdata = r.json()
+                text = rdata["choices"][0]["message"]["content"]
+                if text:
+                    return {
+                        "summary": text,
+                        "title": info["title"],
+                        "video_id": video_id
+                    }
+        except Exception as e:
+            print(f"⚠️ Ollama summary failed: {str(e)[:200]}")
 
         return {"error": "No AI provider available for summary"}
     except Exception as e:
@@ -1026,7 +995,6 @@ async def get_config():
             "version": "2.5",
             "provider_chain": [
                 "cerebras" if cerebras_client else None,
-                "groq" if groq_client else None,
                 "ollama"
             ],
             "features": {
@@ -1043,7 +1011,6 @@ async def get_config():
 @app.get("/health")
 async def health():
     cerebras_status = "not_configured"
-    groq_status = "not_configured"
     ollama_status = "unknown"
 
     if cerebras_client:
@@ -1057,17 +1024,6 @@ async def health():
         except Exception as e:
             cerebras_status = f"error: {str(e)[:100]}"
 
-    if groq_client:
-        try:
-            groq_client.chat.completions.create(
-                model=GROQ_MODELS["fast"],
-                messages=[{"role": "user", "content": "Say OK"}],
-                max_tokens=5,
-            )
-            groq_status = "connected"
-        except Exception as e:
-            groq_status = f"error: {str(e)[:100]}"
-
     try:
         async with httpx.AsyncClient(timeout=5.0) as hc:
             r = await hc.get(f"{OLLAMA_BASE_URL}/models")
@@ -1089,7 +1045,6 @@ async def health():
         "message": "✅ Cortexa Backend Running",
         "provider_chain": [
             {"name": "cerebras", "status": cerebras_status, "primary": True},
-            {"name": "groq", "status": groq_status, "primary": False},
             {"name": "ollama", "status": ollama_status, "primary": False},
         ],
         "youtube_api": "configured" if YOUTUBE_API_KEY else "not configured",
@@ -1108,12 +1063,6 @@ async def list_models():
             "enabled": cerebras_client is not None,
             "primary": True,
         },
-        "groq": {
-            "models": GROQ_MODELS,
-            "preferred": GROQ_PREFERRED,
-            "enabled": groq_client is not None,
-            "primary": False,
-        },
         "ollama": {
             "model": OLLAMA_MODEL,
             "base_url": OLLAMA_BASE_URL,
@@ -1128,7 +1077,7 @@ async def root():
         "message": "🚀 Cortexa Backend Running",
         "version": "2.5",
         "status": "active",
-        "provider_chain": ["cerebras", "groq", "ollama"],
+        "provider_chain": ["cerebras", "ollama"],
         "endpoints": [
             "/api/chat", "/chat",
             "/api/register", "/api/login", "/api/logout", "/api/me",
@@ -1168,7 +1117,6 @@ if __name__ == "__main__":
     print(f"🌐 Port: {port}")
     print(f"💾 DB: {DB_PATH}")
     print(f"🧠 Cerebras: {'✅ configured' if cerebras_client else '❌ missing'}")
-    print(f"🧠 Groq:     {'✅ configured' if groq_client else '⏭️  skipped'}")
     print(f"🦙 Ollama:   {OLLAMA_BASE_URL}")
     print("================================================")