Repeated Gateway Timeout errors with deepseek-3.1

AI & Data Science NVIDIA NIM Models
,
1

Using the model deepseek-ai/deepseek-v3.1, the API keeps returning:

request failed with status 504 Gateway Timeout

This has been ongoing for ~8 hours at least.

Trying to interact with the model via the web interface at https://build.nvidia.com/deepseek-ai/deepseek-v3_1 shows an indicator claiming there are hundreds of requests in queue, then fails with a generic “Internal server error” message.

2

Hi Straddle,

Can you let us know if you are still experiencing the error?

We had an outage, but it should be back up now.

Thanks!