Independent LLM API monitoring
Inference Pulse
Inference Pulse monitors real-world LLM provider API availability, time to first token, throughput, blended price, and 24-hour history across models and regions.
Signals
Compare availability, latency, TTFT, TPS, blended price, and hourly status history for public AI model provider endpoints.
Method
Measurements come from real inference requests to public chat completions endpoints, not provider health checks or synthetic pings.
Scope
Results are a rolling 24-hour best-effort signal from monitored regions. They are independent measurements, not an official SLA.
Live provider status tables load when JavaScript is available. Read the methodology Contact Inference Pulse