# Batch processing custom\_id issue **URL:** \ **Category:** Bugs\ **Created:** [July 18, 2024, 8:53pm UTC](https://community.openai.com/t/batch-processing-custom-id-issue/871808 "2024-07-18T20:53:44Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![alp.ecevit](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/alp.ecevit/32/448292_2.png) [@alp.ecevit](https://community.openai.com/u/alp.ecevit)\ **Post date:** [July 18, 2024, 8:53pm UTC](https://community.openai.com/t/batch-processing-custom-id-issue/871808/1 "2024-07-18T20:53:44Z")
I recently tried batch processing with gpt-4-turbo. I created my request objects and used uuid (specifically uuid4) to generate custom\_id for each of my request. After processing, in the output file, custom\_id fields are totally irrelevant to those I created. It seems like the api generated its own uuids and replaced them with my custom\_id field values. Is this behavior normal? ---
**Author:** ![alp.ecevit](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/alp.ecevit/32/448292_2.png) [@alp.ecevit](https://community.openai.com/u/alp.ecevit)\ **Post date:** [July 18, 2024, 9:28pm UTC](https://community.openai.com/t/batch-processing-custom-id-issue/871808/2 "2024-07-18T21:28:03Z")
Actually, this happens with any custom id. I tried setting the custom\_ids to simple strings like “request-1”, “request-2”, etc. But the returned custom\_ids are newly generated uuids. This prevents me from matching inputs with outputs. ---
**Author:** ![umair313](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/umair313/32/282815_2.png) [@umair313](https://community.openai.com/u/umair313)\ **Post date:** [October 26, 2024, 6:41pm UTC](https://community.openai.com/t/batch-processing-custom-id-issue/871808/3 "2024-10-26T18:41:21Z")
I am also having this issue on batch API output file replaced my custom\_id’s which are UUID’s, now I cant map the response with original one? is there any solution to that.