Status

Live status.

Measured from real requests, never typed in: whether each model is serving, its availability over the last day, and how quickly answers start — the median time to the first token.

Sovereign route

Sovereign route
ModelEndpointNowAvailability · 24hFirst token · median 24h
command-a-03-2025KSA
Chat
Serving
—
—
gpt-oss-120bKSA
Chat
Serving
—
—
gpt-oss-20bKSA
Chat
Serving
—
—
gpt-oss-safeguard-20bKSA
Chat
Serving
—
—
qwen3.8-27bKSA
Chat
Serving
—
—
prompt-guard-2-22mKSA
Chat
Serving
—
—
prompt-guard-2-86mKSA
Chat
Serving
—
—
whisper-large-v3KSA
Speech to text
Serving
—
—
whisper-large-v3-turboKSA
Speech to text
Serving
—
—
orpheus-arabic-saudiKSA
Text to speech
Serving
—
—
orpheus-englishKSA
Text to speech
Serving
—
—

International route

International route
ModelEndpointNowAvailability · 24hFirst token · median 24h
gemini-4-carbon
Chat
Serving
—
—
gemini-3.1-pro-preview
Chat
Serving
—
—
gemini-4-argon
Chat
Serving
—
—
claude-opus-4-6
Chat
Serving
—
—
claude-sonnet-4-6
Chat
Serving
—
—
claude-haiku-4-5
Chat
Serving
—
—
gemini-3.8-flash
Chat
Serving
—
—

Generated 2026-11-21 04:43:17 UTC · /api/status

Availability counts only failures on our side — an endpoint that did not answer, or a full fleet. A request refused for credit or a rate limit is a correct answer and is not counted against us. A dash means no traffic, not perfection.

Build on it.

Each route has its own base URL, models and prices.