Hi,
The black-forest-labs/flux.1-dev model has been unavailable for me for more than 18 hours, both from the API and from the try-it page on build.nvidia.com.
Since about Oct 2, 2026 at 17:00 UTC:
- API (https://ai.api.nvidia.com/v1/genai/black-forest-labs/flux.1-dev): every request times out after 60-120 seconds with no response. Occasionally it returns HTTP 500 “Internal Server Error”. The same requests worked fine before, taking about 6 seconds.
- Web playground (https://build.nvidia.com/black-forest-labs/flux_1-dev): generating an image returns this error:
{
“type”: “about:blank”,
“title”: “Internal Server Error”,
“status”: 500,
“detail”: “Retries exhausted: 3/3”,
“instance”: “/v2/predict/models/qc69jvmznzxy/flux_1-dev”,
“properties”: {
“cause”: “Retries exhausted: 3/3”
}
}
flux.1-schnell also times out. Other endpoints on my account, such as the LLM chat completions at integrate.api.nvidia.com, are working normally, so it doesn’t look like an API key or account problem.
Request example (API key omitted):
POST /v1/genai/black-forest-labs/flux.1-dev
{“prompt”: “a dark hallway at night”, “mode”: “base”, “steps”: 30, “seed”: 12345, “width”: 768, “height”: 1344}
Could you please check the status of the FLUX.1-dev backend? Thanks!