System Status

ModelAPI Status

Endpoint probes refresh every 30s. Model bars are illustrative until the probe series ships — see /trust and /status/history.

Running Checks…

Global Endpoints

— any endpoint has full API access
GL
Global (Primary gateway)PRIMARY
Primary · nearest PoP
api.aimodelapi.ai/v1
Checking…
US
Virginia
Dedicated hostname planned · global host already routes to US East
us.api.aimodelapi.ai/v1
Checking…
SG
Singapore
Asia Pacific PoP
sg.api.aimodelapi.ai/v1
Checking…
JP
Tokyo
Japan PoP
jp.api.aimodelapi.ai/v1
Checking…
TH
Bangkok
Southeast Asia
th.api.aimodelapi.ai/v1
Checking…
EU
Frankfurt
Europe PoP
eu.api.aimodelapi.ai/v1
Checking…
BK
Alt Primary Domain
Backup if .ai is blocked
api2.aimodelapi.ai/v1
Checking…

If the primary is unreachable, swap BASE_URL to any backup (same API key):

# primary
BASE_URL = "https://api.aimodelapi.ai/v1"
# backups (same key, same code)
BASE_URL = "https://sg.api.aimodelapi.ai/v1" # Singapore

AI Provider Status

— illustrative until provider health API ships
OpenAI
All models nominal
Checking…
Anthropic
All models nominal
Checking…
Google
All models nominal
Checking…
DeepSeek
All models nominal
Checking…
xAI
All models nominal
Checking…
Meta
All models nominal
Checking…
Alibaba
All models nominal
Checking…
Moonshot
All models nominal
Checking…

Model Status

— illustrative bars, not audit data
claude-haiku-4-5
ANTHROPIC
Operational
92.76%
Uptime·7d
954ms
Latency
14m ago
Last check
claude-opus-5
ANTHROPIC
Operational
91.45%
Uptime·7d
1800ms
Latency
14m ago
Last check
claude-sonnet-5
ANTHROPIC
Operational
90.13%
Uptime·7d
1100ms
Latency
14m ago
Last check
gpt-5.6-sol
OPENAI
Operational
98.2%
Uptime·7d
820ms
Latency
14m ago
Last check
deepseek-v4-flash
DEEPSEEK
Operational
97.5%
Uptime·7d
520ms
Latency
14m ago
Last check
qwen3.8-max
QWEN
Operational
97.1%
Uptime·7d
620ms
Latency
14m ago
Last check

These model bars are illustrative and cannot be used for SLA or external audit. Method: /status/history · /performance

Incident History

Elevated latency on US East endpoint
2026-05-08Resolved

Resolved in 18 minutes. Root cause: upstream network maintenance at US East gateway. All other endpoints unaffected.

OpenAI API intermittent 503 errors
2026-04-29Resolved

OpenAI upstream degraded for ~40 minutes. ModelAPI auto-failover routed affected requests to alternative model groups. Resolved.

Subscribe to Status Updates

Get email or Telegram alerts the moment any endpoint or provider shows degradation.

Telegram ·Twitter / X