expanded example.env with new LLM model, increased MAX_ARTICLE_LENGTH, updated context size, and extended .gitignore entries

.gitignore

@@ -58,3 +58,8 @@ htmlcov/
 !.yarn/releases
 !.yarn/sdks
 !.yarn/versions
+backend-rust/owlynews.sqlite3
+backend-rust/target
+/backend-rust/config.toml
+/backend-rust/owlynews.sqlite3-shm
+/backend-rust/owlynews.sqlite3-wal

backend/app/config.py

@@ -12,7 +12,7 @@ LLM_MODEL = os.getenv("LLM_MODEL", "mistral-nemo:12b")
 LLM_TIMEOUT_SECONDS = int(os.getenv("LLM_TIMEOUT_SECONDS", 180))
 OLLAMA_API_TIMEOUT_SECONDS = int(os.getenv("OLLAMA_API_TIMEOUT_SECONDS", 10))
 ARTICLE_FETCH_TIMEOUT = int(os.getenv("ARTICLE_FETCH_TIMEOUT", 30))
-MAX_ARTICLE_LENGTH = int(os.getenv("MAX_ARTICLE_LENGTH", 5000))
+MAX_ARTICLE_LENGTH = int(os.getenv("MAX_ARTICLE_LENGTH", 10_000))
 
 frontend_path = os.path.join(
     os.path.dirname(os.path.dirname(os.path.dirname(__file__))),

backend/app/main.py

@@ -15,6 +15,8 @@ import sqlite3
 import time
 from datetime import datetime, timedelta, timezone
 from typing import Any, Dict, List, Union
+import subprocess
+import threading
 
 # Third-party imports
 import httpx
@@ -62,6 +64,19 @@ scheduler.add_job(
 scheduler.start()
 
 
+def start_frontend_build():
+    try:
+        subprocess.Popen(
+            ["yarn", "build"],
+            cwd="../frontend",
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+        )
+        print("Frontend build started successfully")
+    except Exception as e:
+        print(f"Failed to start frontend build: {e}")
+
+
 # API endpoints
 @app.get("/news", response_model=List[Dict[str, Any]])
 async def get_news(
@@ -200,8 +215,6 @@ async def list_feeds(db: sqlite3.Cursor = Depends(get_db)):
         )
 
 
-
-
 @app.post("/feeds", response_model=SuccessResponse)
 async def add_feed(
         feed: FeedData,
@@ -332,7 +345,6 @@ async def manual_sync(db: sqlite3.Cursor = Depends(get_db_write)):  # Note: chan
         )
 
 
-
 @app.get("/meta/last-sync", response_model=TimestampResponse)
 async def get_last_sync(db: sqlite3.Cursor = Depends(get_db)):
     """
@@ -406,5 +418,11 @@ async def update_cron_schedule(
     return {"hours": hours}
 
 
-# Mount static frontend
 app.mount("/", StaticFiles(directory=frontend_path, html=True), name="static")
+
+if __name__ == "__main__":
+    threading.Thread(target=start_frontend_build, daemon=True).start()
+
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=8000)
+

backend/app/services.py

@@ -209,7 +209,7 @@ class NewsFetcher:
             "format": "json",
             "options": {
                 "num_gpu": 1,  # Force GPU usage
-                "num_ctx": 8192,  # Context size
+                "num_ctx": 128_000,  # Context size
             }
         }
 

backend/example.env

@@ -14,6 +14,7 @@ SYNC_COOLDOWN_MINUTES=30
 LLM_MODEL=qwen2:7b-instruct-q4_K_M
 LLM_MODEL=phi3:3.8b-mini-128k-instruct-q4_0
 LLM_MODEL=mistral-nemo:12b
+LLM_MODEL=cnjack/mistral-samll-3.1:24b-it-q4_K_S
 
 # Timeout in seconds for LLM requests
 LLM_TIMEOUT_SECONDS=180