High Demand Error

[cheang150]

What version of Codex is running?

gpt-5.1

What subscription do you have?

Enterprise

Which model were you using?

gpt-5.1

What platform is your computer?

Microsoft Windows NT 10.0.19045.0 x64

What issue are you seeing?

Configured ~/.codex/config.toml as below

model = "gpt-5.1"
model_provider = "nwg"

[model_providers.nwg]
name = "gpt-5.1"
base_url = "https://nwgateway-appdev.kepler-prod.shared.banksvcs.net/genai/openai/gpt51_dzstd/openai/deployments/gpt51_dzstd_prod_sec_2025-11-13"
wire_api = "chat"
env_key = "OPENAI_API_KEY"
query_params = { api-version = "2024-12-01-preview" }

Just a simple codex "hello" get me We're currently experiencing high demand, which may cause temporary errors.
The status started with Working, then Reconnecting and finally the high demand error. I've tried to eliminate the possibility of custom endpoint issue by running curl and both curl and codex is using the same JWT token.

PS C:\dev\workspace> curl.exe -X POST `
>>   "https://nwgateway-appdev.kepler-prod.shared.banksvcs.net/genai/openai/gpt51_dzstd/openai/deployments/gpt51_dzstd_prod_sec_2025-11-13/chat/completions?api-version=2024-12-01-preview" `
>>   -H "Content-Type: application/json" `
>>   -H "Authorization: Bearer $env:OPENAI_API_KEY" `
>>   -d '{\"messages\": [{\"role\": \"user\", \"content\": \"hello\"}]}'
{"choices":[{"content_filter_results":{},"finish_reason":"stop","index":0,"logprobs":null,"message":{"annotations":[],"content":"Hello! How can I help you today?","refusal":null,"role":"assistant"}}],"created":1765962186,"id":"chatcmpl-Cnhj04
KURALRdjqUb6XFp4MbKKhJc","model":"gpt-5.1-2025-11-13","object":"chat.completion","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{}}],"system_fingerprint":null,"usage":{"completion_tokens":19,"completion_tokens_details":{"
accepted_prediction_tokens":0,"audio_tokens":0,"reasoning_tokens":0,"rejected_prediction_tokens":0},"prompt_tokens":7,"prompt_tokens_details":{"audio_tokens":0,"cached_tokens":0},"total_tokens":26}}

I have tried to restart codex session as well, the error seems to be consistent.

What steps can reproduce the bug?

Uploaded thread: 019b2b85-474e-7a32-b144-d80ceefc05c4

What is the expected behavior?

Codex should run fine without error

Additional information

I am aware of #8120 but it seems to be closed as resolved as fix has been deployed 9 hours ago. But this issue is still persistent for me.

[github-actions]

Potential duplicates detected. Please review them and close your issue if it is a duplicate.

• Experiencing high demand, which may cause temporary errors #8135
• We're currently experiencing high demand, which may cause temporary errors. #8130
• We're currently experiencing high demand, which may cause temporary errors. #6761

Powered by Codex Action

[etraut-openai]

It looks like you're using some sort of an LLM proxy server. Based on the symptoms you're describing, it's likely that the proxy server is the cause of this problem. The codex error message that you're seeing is shown when it receives a 500 error from the responses API endpoint. We monitor 500 errors closely on our end, and while they do occasionally occur, it's unlikely that any single user would see 500 errors repeatedly. I recommend working with your IT organization or whoever runs your LLM proxy server.

I wasn't able to get any additional diagnostic information because I couldn't find any logs or session details for the thread ID that you listed in your bug report. Did you use the /feedback tool to upload it?