Is Fireworks AI down?

Fireworks AI

No — Fireworks AI is working normally

Our sources see nothing wrong. If it still fails for you, the cause is most likely on your side — the tips further down cover the usual suspects.

Working since · Checked at

What our sources say

Official status page
Working
Our test requests, last hour
Not running for this service
Traffic through OpenRouter
0 of 4 models failing, 100% of requests succeeded in 30 min
Reports from visitors, last hour
0

Parts of the service

Official status page ↗Open Fireworks AI ↗

Last 24 hours

16:46 UTC — Working17:16 UTC — Working17:46 UTC — Working18:16 UTC — Working18:46 UTC — Working19:16 UTC — Working19:46 UTC — Working20:16 UTC — Working20:46 UTC — Working21:16 UTC — Working21:46 UTC — Working22:16 UTC — Working22:46 UTC — Working23:16 UTC — Working23:46 UTC — Working00:16 UTC — Working00:46 UTC — Working01:16 UTC — Working01:46 UTC — Working02:16 UTC — Working02:46 UTC — Working03:16 UTC — Working03:46 UTC — Working04:16 UTC — Working04:46 UTC — Working05:16 UTC — Slow or glitchy05:46 UTC — Slow or glitchy06:16 UTC — Working06:46 UTC — Slow or glitchy07:16 UTC — Slow or glitchy07:46 UTC — Slow or glitchy08:16 UTC — Working08:46 UTC — Working09:16 UTC — Working09:46 UTC — Working10:16 UTC — Working10:46 UTC — Working11:16 UTC — Working11:46 UTC — Working12:16 UTC — Working12:46 UTC — Working13:16 UTC — Working13:46 UTC — Working14:16 UTC — Working14:46 UTC — Slow or glitchy15:16 UTC — Slow or glitchy15:46 UTC — Working16:16 UTC — Working

24 h agonow

Last 30 days

28 Sept — No data29 Sept — No data30 Sept — No data1 Oct — No data2 Oct — No data3 Oct — No data4 Oct — No data5 Oct — No data6 Oct — No data7 Oct — No data8 Oct — No data9 Oct — No data10 Oct — No data11 Oct — No data12 Oct — No data13 Oct — No data14 Oct — Slow or glitchy15 Oct — Working16 Oct — Slow or glitchy17 Oct — Slow or glitchy18 Oct — Slow or glitchy19 Oct — Slow or glitchy20 Oct — Slow or glitchy21 Oct — Slow or glitchy22 Oct — Slow or glitchy23 Oct — Working24 Oct — Working25 Oct — Slow or glitchy26 Oct — Slow or glitchy27 Oct — Slow or glitchy

30 days agoyesterday

3 of 14 days with data had no problems.

Recent incidents

  1. Slow or glitchy

    Service Degradation for GLM 5.3 Flash

    Resolved after 8 mins

  2. Slow or glitchy

    Service Degradation for GLM 5.3 Flash

    Resolved after 57 mins

  3. Slow or glitchy

    Service Degradation for MiniMax M3.1 Flash

    Resolved after 2 mins

  4. Slow or glitchy

    Service Degradation for MiniMax M3.1 Flash

    Resolved after 3 mins

  5. Slow or glitchy

    Service Degradation for MiniMax M3.1 Flash

    Resolved after 4 mins

  6. Slow or glitchy

    Service Degradation for GLM 5.3 Flash

    Resolved after 24 mins

  7. Slow or glitchy

    Service Degradation for Kimi K3

    Resolved after 30 mins

  8. Slow or glitchy

    Service Degradation for Kimi K3

    Resolved after 7 mins

About this service

Fireworks AI serves open models and fine-tunes. Serverless models and dedicated deployments fail independently, so a serverless incident may never touch your dedicated endpoint.

Other model APIs right now

Fireworks AI: common questions

Is Fireworks AI down, or just one model?

Fireworks AI serves many models, and they rarely fail all together. If one model returns errors, try another of a similar size: if that works, the trouble is limited to that model or the hardware it runs on. When unrelated models fail at the same moment, the platform itself is struggling.

What can I do while Fireworks AI has problems?

Most model hosts speak the OpenAI-compatible API, so moving traffic to another host that serves the same open model is often just a different base URL and key. Set that fallback up before you need it, and meanwhile retry 5xx errors with growing pauses.

Why am I getting 429 errors from Fireworks AI?

A 429 means you've reached a request, token or concurrency limit of your plan, not that the service is down. Limits differ from model to model: spread requests out, upgrade the plan or pick a model with more headroom.

Why does it work for a colleague and not for me?

Requests can land on different hardware depending on the account and the model, so two people can see different results in the same minute. Your key's limits, the exact model ID and the region you call from all play a part.

How do you know whether Fireworks AI is down?

For Fireworks AI we combine Fireworks AI's official status page, read every minute, the share of successful requests to Fireworks AI's models on OpenRouter and reports from people who press the button on this page. The most serious signal that is still fresh sets the verdict, and stale sources are ignored rather than trusted. The page refreshes by itself; the exact thresholds are on our “How we check” page.

How we check →