Requesty
Live Status

LLM Provider Status

Real-time status for every major LLM provider. Live data from OpenAI, Anthropic, and Cohere — updated every 60 seconds.

Live Monitoring

OpenAI

Loading...
api.openai.comFull status

Anthropic

Loading...
api.anthropic.comFull status

Cohere

Loading...
api.cohere.aiFull status

Other Providers

Google (Vertex AI)

Status page
generativelanguage.googleapis.com
Gemini 2.5 Pro, Gemini 2.5 Flash
Cloud-grade SLA. Uses Google Cloud infrastructure.

DeepSeek

Status page
api.deepseek.com
DeepSeek V3, DeepSeek R1
Lowest pricing. Occasional latency spikes under heavy load.

Mistral

Status page
api.mistral.ai
Mistral Large, Mistral Small, Codestral
EU-based provider. GDPR compliant by default.

Groq

Status page
api.groq.com
Llama 4 Scout, Llama 3.3 70B
Fastest inference (custom LPU hardware). Limited model selection.

Together AI

Status page
api.together.xyz
Llama, Mixtral, Qwen, open-source models
Open-source model hosting. Good for Llama and Mixtral variants.

AWS Bedrock

Status page
bedrock-runtime.*.amazonaws.com
Claude, Llama, Mistral (via AWS)
Enterprise SLA. Models accessed through AWS account.

Azure OpenAI

Status page
*.openai.azure.com
GPT-4, GPT-4o (via Azure)
Enterprise SLA. Dedicated capacity options available.

Never go down because a provider does

Multi-provider failover gives your application composite uptime higher than any single provider. Requesty detects failures in under 100ms and reroutes automatically.

Automatic failover

Define fallback chains. When Provider A fails, traffic instantly moves to Provider B. No code changes.

99.99% uptime SLA

Requesty's infrastructure is multi-region. Combined with multi-provider routing, your effective uptime exceeds any single provider.

Health monitoring

Real-time health checks on every provider. Route away from degraded endpoints before they fail completely.

Protect your app from outages

One API, 400+ models, automatic failover. Start free with $10 credits.

Start free

Frequently asked questions

Is OpenAI down right now?

Check the official OpenAI status page at status.openai.com for real-time updates. OpenAI experiences partial or full outages multiple times per month. For production applications, use an AI gateway with automatic failover so outages do not affect your users. Requesty automatically reroutes to backup providers (Anthropic, Google, DeepSeek) when OpenAI is down.

How do I protect my app from LLM provider outages?

Use an AI gateway with automatic failover. Define a fallback chain of 2 to 4 providers (e.g. OpenAI then Anthropic then Google). The gateway detects failures in under 100ms and reroutes to the next provider. Add prompt caching so cached responses serve even during total outages. Requesty provides all of this out of the box with no code changes needed.

Which LLM provider has the best uptime?

Cloud-hosted options (AWS Bedrock, Azure OpenAI) offer the strongest SLAs at 99.9%+. Among direct API providers, Anthropic and Google tend to have the fewest outages. However, no single provider guarantees 100% uptime. The most reliable approach is multi-provider failover through an AI gateway, which gives you composite uptime higher than any individual provider.