🏺 Pithos Configuration
Customize outbound token reduction filters and parameters.
License & Integration
License Key
Free
Activate
Enter your license key to unlock Pro or Team capabilities.
Local Proxy Port
coming soon
Copy
Paste this as your AI tool's base URL to proxy traffic.
Optimization Settings
Output Minimization Mode
Diff Only
Restricts output strictly to standard git diff files. Saves up to 80% on outputs.
Concise Explanation
Generates only essential context alongside edits. Standard coder balance.
Full Output
Disable output formatting constraint rules. Retains verbose answers.
Max Context Size
4000
Sets the maximum allowed context files buffer size per request.
Local Evaluation Model
paraphrase-multilingual-MiniLM-L12-v2
Local embedding model (bundled) used to evaluate context files.
Smart Model Routing
Route easy tasks to cheaper models automatically to reduce cost.
Semantic Cache
Reuse previous optimization results for identical or repeated tasks.
System Status
Local Middleware Connection
Connected
✓
Privacy Guarantee:
All prompt evaluation checks occur offline on your localhost. Your files never pass through remote cloud wrapper relays.
Save Settings
Reset to Defaults