Common issues with OpenAI Chat Model in Duaer

Expressions resolve to the first item

OpenAI Chat Model is a sub-node. Most nodes resolve an expression once per input item; Duaer calls a sub-node with the first item only. Given five names, {{ $json.name }} is always the first name.

To use a different value per item, put the expression in the root node (such as the AI Agent prompt), or run Loop Over Items first so each run gets one item.

This model requires the Responses API

The selected model only works with the Responses API. Turn on Use Responses API, or pick a model that supports Chat Completions.

Insufficient quota detected

The OpenAI account has no credit or is over a usage limit. The Duaer OpenAI Chat Model shows this on insufficient_quota. In the OpenAI console:

1. Pick the organization and project that own the API key in the upper-left selectors.
2. Check that the organization Billing page shows credit. A fresh top-up can take a while to apply.
3. Check organization Limits and project Limits for a usage cap you have reached.

Back in Duaer, confirm the OpenAI credential holds a key from the funded account and that the node uses that credential. If the error persists after a top-up, create a new key and try again.

OpenAI: Rate limit reached

OpenAI returned 429 because of too many requests in a short time.

Duaer retries up to Options > Max Retries (default 2). If you still hit the limit, lower Batch Size and raise Delay Between Batches under Batch Processing on the root node.

Bad request - please check your parameters

OpenAI rejected the request and Duaer could not read the reason. Check the model name and parameters first. Then send the same request from HTTP Request with the same credential to see the raw OpenAI error.