Certains systèmes sont dégradés
11/12
fournisseurs en ligne
DeepSeek
Degraded
890mslatence
97.42%uptime 30j
High latency on all models
api.deepseek.com
OpenAI
Operational
142mslatence
99.87%uptime 30j
api.openai.com
Anthropic
Operational
98mslatence
99.95%uptime 30j
api.anthropic.com
Google AI
Operational
115mslatence
99.91%uptime 30j
generativelanguage.googleapis.com
Groq
Operational
34mslatence
99.72%uptime 30j
api.groq.com
Together AI
Operational
167mslatence
99.68%uptime 30j
api.together.xyz
Fireworks AI
Operational
89mslatence
99.84%uptime 30j
api.fireworks.ai
Mistral AI
Operational
124mslatence
99.89%uptime 30j
api.mistral.ai
xAI
Operational
203mslatence
99.51%uptime 30j
api.x.ai
Cohere
Operational
156mslatence
99.93%uptime 30j
api.cohere.com
AWS Bedrock
Operational
178mslatence
99.99%uptime 30j
bedrock-runtime.us-east-1.amazonaws.com
Cerebras
Operational
21mslatence
99.78%uptime 30j
api.cerebras.ai
Incidents récents
majorinvestigating
High latency on all models
DeepSeek — Mar 26, 12:15 AM
majorresolved
Elevated API error rates
OpenAI — Mar 24, 02:30 PM — 2h 15m
minorresolved
Rate limiting issues
Groq — Mar 22, 07:00 PM — 1h
minorresolved
Grok 3 intermittent failures
xAI — Mar 21, 03:00 PM — 1h 30m
minorresolved
GPT-4.1 timeout issues
OpenAI — Mar 20, 08:00 AM — 1h 30m
minorresolved
Increased latency on Claude Opus
Anthropic — Mar 18, 11:00 AM — 1h 15m
criticalresolved
API 503 errors
DeepSeek — Mar 15, 06:00 AM — 8h
Vérifié toutes les 60 secondes. Latence mesurée depuis US-East.
Les données reflètent la santé des endpoints API, pas la disponibilité individuelle des modèles.
Les données reflètent la santé des endpoints API, pas la disponibilité individuelle des modèles.