Token & Request Activity
Input, cached, output & reasoning token breakdown over time
Input
Cached
Output
Reasoning
Model Distribution
Only the 3 curated Surplus models
Top Employee Keys
Usage, cache rate & spend per key
Recent Errors & Anomalies
Click any row to inspect the full prompt, headers & upstream error payload
Live Proxy Playground
Send a test prompt through SurplusGate and inspect live telemetry
Coding Harness Integration
Drop-in OpenAI-compatible configuration for any agent or IDE
Gateway Policies
Control model allowlisting and stream capture granularity
Enforce Curated 3-Model Allowlist
Block inference requests for any model outside GLM 5.3, GLM 5.3 Flash, and DeepSeek V4.1 Flash 1017.
Store Raw SSE Stream Chunks
Capture raw verbatim SSE chunk payloads alongside reconstructed responses for deep stream debugging.