About Gemini API Status
Google Gemini (formerly Bard / PaLM) provides the Gemini Pro, Flash, and Ultra models via Google AI Studio and the Vertex AI platform. This page tracks Gemini API outages, degradations, and incidents in real time, automatically updated every 60 seconds from our monitoring infrastructure.
Official status page: https://status.cloud.google.com
Common Gemini Outage Symptoms
- ✕RESOURCE_EXHAUSTED (429) — quota exceeded on free tier or project-level rate limits
- ✕UNAVAILABLE (503) — Vertex AI or Gemini API endpoint unavailable
- ✕Long first-token latency on Gemini Pro and Gemini Flash models
- ✕Quota errors that don't clearly indicate which limit (RPM, TPM, RPD) was hit
- ✕Project-level quota inconsistencies across GCP regions
- ✕Streaming chunks arriving out of order or with unexpected gaps
What to Do During a Gemini Outage
- Check your project-level quota in the Google Cloud Console — Gemini errors often reflect project quotas, not global outages.
- Use exponential backoff with jitter on RESOURCE_EXHAUSTED: start at 1 s, cap at 60 s.
- Distribute load across multiple GCP regions (us-central1, us-east4, europe-west4) if you see region-specific issues.
- Switch to a BYOK proxy (AI Badgr) to get per-request receipts and automatic retry handling across regions.
- Monitor Google Cloud status at status.cloud.google.com — filter for 'Vertex AI' and 'Generative AI on Vertex AI'.
Other AI Provider Status Pages
Gemini Outage FAQ
Is Google Gemini / Vertex AI down right now?
This page checks our live monitoring infrastructure (updated every 60 s) which tracks the Google Cloud incidents feed and our own request telemetry. The status badge at the top reflects the current state.
Why am I getting Gemini RESOURCE_EXHAUSTED errors?
RESOURCE_EXHAUSTED maps to HTTP 429 and means you've exceeded a quota. Google Gemini has per-project rate limits (RPM, TPM, RPD) that are lower on the free tier. Check your quota page in Google Cloud Console and request an increase if needed.
What is the difference between a Gemini outage and a quota error?
An outage affects all users globally — it appears on status.cloud.google.com. A quota error is specific to your project. If only your project sees errors and Google's status page is green, it's a quota issue, not an outage.
Can I automatically failover away from Gemini during an outage?
Yes. AI Badgr can route Gemini-compatible requests to alternate backends automatically. Change one configuration line and get transparent receipts for every request, plus automatic retry and failover logic.
How do I monitor Gemini / Vertex AI incidents?
Subscribe to Google Cloud status at status.cloud.google.com and filter for Vertex AI. AI Badgr also monitors the public incidents feed and can send you an email alert when a degradation is detected.
Never get stuck in a Gemini outage again
AI Badgr acts as a transparent proxy for your existing API keys. One line of code change. Zero vendor lock-in. Instant failover when Gemini is down.
Get Started Free →