The free tier is closed for most users.Thank you for every chat along the way. Paid models work as before, and your account, API keys, and balance stay as they are.Read why

All systems operational

Updated 12:39:11 AM UTCnext update in 30s

Overall uptime

100.00%

Last 24 hours
100.00%
Last 7 days
100.00%
Last 30 days
100.00%

Services Last 90 days

  • Website Up100.00% uptime
  • API Up100.00% uptime
  • Marketplace models Up100.00% uptime

Each bar is one day of checks. The dot and word show the state right now. A healthy day is a full green bar; a worse day is shorter.

  • 99.9% or more
  • 99% to 99.9%
  • 95% to 99%
  • Under 95%
  • No data

Provider health

Live health of every provider, measured on real requests through the FreeTheAI gateway.

Providers up
4/4
Models listed
9
Last check
12:39:11 AM UTC

Each bar is one day of real requests (one hour inside a provider). A healthy day is a full green bar; a day with more failed requests is shorter.

  • 95% or more
  • 80% to 95%
  • Under 80%
  • No data
  • DuskyLotusDuskyLotus Up4 of 6 models up
    78.03%uptime, 7 days
    • claude-sonnet-5.5claude-sonnet-5.5
      100.00%85.59%85.59%85.59%
    • glm-5.3glm-5.3
      100.00%100.00%100.00%100.00%
    • glm-5.3-flashglm-5.3-flashNo requests in the last 24 hours—100.00%100.00%100.00%
    • gpt-oss-120b-freegpt-oss-120b-freeNo requests in the last 24 hours—0.00%0.00%0.00%
    • kimi-k3kimi-k3
      100.00%33.33%33.33%33.33%
    • minimax-m2.5-freeminimax-m2.5-freeNo requests in the last 24 hours—42.86%42.86%42.86%
  • WanderingBrookWanderingBrook Up1 of 1 model up
    96.49%uptime, 7 days
  • WhiteSummitWhiteSummit Up0 of 1 model up
    0.00%uptime, 7 days
  • WildIslandWildIsland Up1 of 1 model up
    100.00%uptime, 7 days
    • claude-sonnet-5.5claude-sonnet-5.5No requests in the last 24 hours—100.00%100.00%100.00%

Open a provider to see each model. A dash means no requests in that window.

Status updates

No incidents in the last 30 days.

How status is decided

Uptime is checked once a minute from our servers through the public internet: the website and API must answer, and the model groups must have at least one provider serving. Two failed checks in a row open an incident; the next good check closes it.

Status looks at the last 30 minutes of requests the provider answered or failed. A provider is down when it has no models, when 3 or more requests all failed, or when 90% or more of at least 5 failed; degraded when 25% or more of at least 5 failed; otherwise up. A caller's own bad request never counts against a provider. Uptime is the share of successful requests.