Hello,
I’m testing out the GenerativeAIExamples and just came across an issue/question.
While running the multimodel_rag example local_deploy I am getting the following error: " Cloud credits expired - Please contact NVIDIA representatives ERROR:RAG.src.chain_server.server:Error from POST /documents endpoint".
I currently do not have any cloud credits so I understand the error. I am just unsure why it’s trying to use cloud credits when I am running everything locally/on-prem. What am I missing?
Thanks!
Hi,
The error occurs because, even in a local deployment, the multimodel_rag pipeline still requires calling endpoints like google/deplot and nvidia/neva-22 from the API Catalog. These external endpoints need cloud credits to function, even if you’re running other components locally.
You can sign up for a free developer license to access credits here:
Hope this helps!
Exception: [402] Payment Required
Account ‘S_dxr6YUEBIUiB3QwTp-E3H74ZsnFG_KFgcW0FD6_lI’: Cloud credits expired - Please contact NVIDIA representatives
get the same error
openai.APIStatusError: Error code: 402 - {‘status’: 402, ‘title’: ‘Payment Required’, ‘detail’: “Account ‘EUBDRse2hERIe52kemvgdw4mDDheQZrwkcdj60qC9ME’: Cloud credits expired - Please contact NVIDIA representatives”}
But I still have 1000 credits, so what could be the problem?
I try to use these models:
model_dict = {
“nvidia - nemotron-mini-4b”: “nvidia/nemotron-mini-4b-instruct”,
“ibm - granite-8b”: “ibm/granite-3.0-8b-instruct”,
“zyphra - zamba2-7b”: “zyphra/zamba2-7b-instruct”,
“meta - llama-8b”: “meta/llama-3.1-8b-instruct”}