DownForAI

Perplexity status: API, auth, latency & outage reports

Unknown
Last probe 5 min ago·2 surfaces
Likely your sideMEDIUM confidence
📡 Official status page →📚 Docs →💳 Pricing →
0.0%
24h Uptime
57ms
p50 Latency
111ms
p95 Latency
0
Incidents (30d)

Having issues with Perplexity?

Report problems quickly and help the community stay informed.

0
reports in the last 24 hours

Surface Health

Perplexity APIUnknown
HTTP 403p50 53ms4m ago
Perplexity WebUnknown
HTTP 403p50 60ms22m ago

Uptime — last 24h

0.0%

Latency — last 24h (p50 per 30 min slot)

Loading…

Is Perplexity down for everyone?

Likely local or client-side issue
Our probes see normal responses. The issue is likely on your end or in your network path.
Moderate confidence
Probe summary (2 surfaces)
All surfaces operational as of last probe.
Signals detected
  • All monitored surfaces operational
  • No recent user reports
  • Check your network, credentials, or rate limits

Incident history (30d)

✓No incidents recorded in the past 30 days.

Reported symptoms

No user reports for Perplexity in the last 24 hours.

Known error signatures

Common failure patterns and how to diagnose them

Provider details

AI-powered search engine with sourced answers. Combines LLM reasoning with real-time web search and citations. Consumer app plus Sonar API for developers.

What we monitor
perplexity.ai web — Consumer web interface
Perplexity mobile apps — Mobile backend
Sonar API — Developer API
Web search backend — Scraping/indexing infrastructure (unique dependency)
Status page segments
Web AppAPISearch Infrastructure
Model families
Sonar (proprietary search-augmented)Sonar ProClaude, GPT, Grok (Pro user choice for underlying reasoning)
Common limits & quotas
  • Free tier: limited Pro Searches per day
  • Pro tier ($20/mo): expanded Pro Searches, Deep Research access
  • Sonar API: pay-per-request with tier-based quotas
Ecosystem dependencies
Depends on web scraping/crawling infrastructure (proprietary plus upstream search providers)Underlying LLM models (Claude, GPT, etc.) — cascades possible from those providers
Operator notes
  • Source fetching failures are often upstream (the sites Perplexity is trying to read), not Perplexity itself
  • Deep Research is a long-running task; UI 'stuck' perception is common — timeouts are 5-10 min typically
  • Pro users can select the underlying reasoning model; a model-specific issue can be worked around without leaving Perplexity
  • Real-time news indexing has delay — not a real-time search engine like Google News
Diagnostic signals
$ curl https://api.perplexity.ai/chat/completions -H "Authorization: Bearer $PPLX_API_KEY" -H "Content-Type: application/json" -d '{"model":"sonar","messages":[{"role":"user","content":"ping"}]}' — Sonar API reachability$ Check status.perplexity.ai for component-level breakdown

Fallback alternatives

What to use if this service is down

Perplexity is degraded
You.com or Kagi offer similar AI-powered search
Easy switch
Coding search specifically
Phind is an alternative
Easy switch
Only source retrieval is broken
Direct ChatGPT with web browsing or Claude with web access can reduce downtime
Easy switch
Deep research workloads
Claude Projects with manual document upload can substitute for Perplexity Deep Research
Manual setup
Moderate effort

How we monitor

downforai.com probes each AI service every 2–5 minutes from multiple independent locations. We measure HTTP response codes, latency (p50 & p95), and endpoint availability across the surfaces listed above. Status is classified as Operational, Degraded, or Outage based on a weighted combination of probe results. Uptime is calculated over 30-minute buckets and the last 24 hours. User reports are factored into our diagnosis as a secondary signal. We are independent of all providers listed and receive no compensation to report any particular status.

Community Discussion

0/1000
No comments yet. Be the first to share your experience!