§ Status checking…

All systems ….

Live health of every backend the API depends on. Polled in real time when this page is open.

Azure AI Foundry

checking

Serverless inference for DeepSeek, Llama, Mistral, Kimi, GPT-OSS. 9 of 16 models route here.

checking

GPT-5 and GPT-5.2. Premium proprietary tier.

checking

Self-hosted on our own GPU server fleet. Gemma 3 27B, GLM-4 9B, Qwen 3 8B, Qwen 2.5 Coder, Mistral Nemo, embeddings.

Models

—

Calls (24h)

—

Tokens (24h)

—

Avg latency

—

For incidents and maintenance announcements, follow [email protected]. We email all active accounts before scheduled downtime.