Overview
Your usage, model breakdown, and recent request waterfall.
Available Models & Token Capacity LIVE POOL
Unified model endpoints, active multi-provider pools, and remaining token capacity across healthy keys.
Your requests over time
Daily API request volume across all connected backends.
Model usage distribution
Traffic share across unified models and active backends.
Live Request Waterfall
Real-time execution log showing selected provider, key rotation, and context rehydration.
| Time | Unified Model | Provider Used | Key Secret | Usage Mode | Status | Latency | Rehydrated |
|---|---|---|---|---|---|---|---|
| No requests logged yet. Use Chat Playground or call /v1/chat/completions. | |||||||
Unified Models
All providers advertising the same model collapse into 1 unified model. Choose your routing strategy below.
Custom Providers & Key Pools
Manage your upstream endpoints, test live latencies, and monitor rate limits.
Cross-Session Memory & Compaction
Persistent facts, directives, and automatic rehydration across rotated keys and providers.
Memory Bank Configuration
Store Custom Directive / Fact
Head-Middle-Tail Compaction
Prevents context window overflow while preserving system prompts and recent conversation turns 100% verbatim.
Active Cross-Session Memory Bank
| Session ID | Category | Directive Key | Memory Content | Updated | Actions |
|---|---|---|---|---|---|
| No cross-session memories stored yet. | |||||
Waterfall Telemetry
Audit trail of all request attempts, key failovers, latencies, and quota quarantines.
| Timestamp | Unified Model | Provider | Key Label | Usage Mode | Status | Latency | Compacted | Rehydrated |
|---|---|---|---|---|---|---|---|---|
| No logs recorded yet. | ||||||||
API Integration & Cloud Gateway
Connect any OpenAI-compatible application, coding assistant, agent, or client to Cyrix AI.