Message for Smith from OpenAI Support

Smith, you closed this ticket of mine last week.

You said that you would deal with it through email instead.
You did email once, to tell me that you had “escalated it to specialists”.
But nothing since then!

I have emailed you again multiple times to follow up and you are ghosting me.

I have to now be suspicious of the reasons that you closed the thread for.
Can you PLEASE respond to my…

Read more →
N
Federal Agency Working to Stop Invasive Bugs Invaded by Bugs

Bedbugs have infested the government agency responsible for insect defense.

The pests were found in the building housing the Animal and Plant Health Inspection Service, an agency of the Department of Agriculture, NOTUS reports. A USDA employee told the publication that the irony “was lost on no one.”

Employees of the agency, located in the George Washington Carver Center in Beltsville,…

Read more →
Ben Hutchings: FOSS activity in 2025

This was a particularly busy month for me in terms of Debian contributions.

It started with a week in Hamburg for the MiniDebConf. I talked to many colleagues face-to-face and worked on various bugs and maintenance tasks. I’m pleased to have finally found the time to reproduce and fix the boot-time crashes in the parallel port
subsystem that have been reported many times recently.

A series of…

Read more →
Can't read past chat anymore “This conversation is too long, please start a new one.”

Welcome to the forum!

Hitting the maximum length and being unable to continue the chat is expected, but I would not expect the old conversation to become unreadable or open as a blank screen afterward.

If you haven’t already, I’d try first checking the same conversation from another browser/device and also from ChatGPT web rather than only the app. If it still opens blank everywhere, that…

Read more →
Contact support: Uh oh. Something went wrong. Please try again later

Welcome to the forum!

I’ll recommend you to check this topic since it’s similar:

My account says I need to pay, but the invoice says Void, so I can't pay it, account is unusable Bugs

My account on the API side says I have unpaid invoices, so it wont’ work.. there I don’t have anything unpaid.. on the chatgpt side i see one that says void.. can you please fix it?

J
Tick that causes meat allergy, other rare virus is spreading: What symptoms to watch for

There were more than 110,000 suspected cases of AGS in the U.S. between 2010 and 2022, according to the CDC, though the true number of cases could be as much as 450,000. The syndrome is not a notifiable condition to the agency.

Research suggests the syndrome is not specific to the lone star tick alone. A pair of studies last year found the condition may also be caused by blacklegged ticks…

Vector Store file delete is not removing the file from vector store

That sounds like the same issue, especially since it persists several days later and affects estimated usage.

If both removing the vector store file object and deleting the underlying file object still leave it visible/counting in the vector store. Like I already recommended in the post above, that this seems worth escalating to Support with the vector store IDs, file IDs, timestamps, API…

Read more →
Ben Hutchings: FOSS activity in April 2026
  • Debian packages:
    • firmware-nonfree:
      • Bugs:
        • replied to #1132448: firmware-mediatek: Wifi perf degraded a lot after upgrade firmware-mediatek from 20250410-2 to 20260221-1~bpo13+1 and 20260309-1
      • Merge requests:
        • merged !142: Replace copy-firmware.sh; install files and generate metainfo.xml at build time
        • opened and merged !143: Clean up…
Read more →
Not able to request batch api of GPT 5.2

Was this a status that was immediately received as a response to job submission, or was it the final report of the batch job?

Reproduction?

[1/5] JSONL file written: repro_batch_input.jsonl
      Contents:
      {"custom_id": "repro-request-1", "method": "POST", "url": "/v1/responses", "body": {"model": "gpt-5.2", "max_output_tokens": 64, "input": "Reply with exactly one…
Read more →
Gpt-5.4 ignores reasoning_effort="none" when max_completion_tokens is used

The fault seen here is that Chat Completions does not deliver the output if it is not complete. It is all classified as “reasoning” in usage, even if it is clear the output would have transitioned to the final seen output.

Then, that there actually is reasoning at “none”, just hidden behind a threshold of 128 tokens, before which where it is not billed.

I’ve made posts about this before. Let’s…

Read more →
Ben Hutchings: FOSS activity in March 2026
  • Debian packages:
    • firmware-nonfree:
      • Bugs:
        • closed #1064620: firmware-nonfree: suggestions for the packaging, gencontrol.py and debian/rules
        • closed #1126797: firmware-intel-graphics: Please ship irci_irci_ecr-master_20161208_0213_20170112_1500.bin as ipu3-fw.bin
        • closed #1131751: ABI break in amdxdna npu firmware
      • Merge requests:
        • opened…
Read more →
I enabled data sharing, but the free tokens were not granted.

Related:

Complimentery tokens: GPT-5.4-mini is charged for 1m token group rather than 10m token group Bugs

Hello, As the title says, I get charged for using more than 1M tokens when using GPT-5.4-mini, and this does seem to be shared with GPT-5.4. However, the popup in the settings clearly states: You’re eligible for free daily usage on traffic shared with OpenAI. Up to 1 million tokens per…

Read more →
I enabled data sharing, but the free tokens were not granted.

Related:

Complimentery tokens: GPT-5.4-mini is charged for 1m token group rather than 10m token group Bugs

Hello, As the title says, I get charged for using more than 1M tokens when using GPT-5.4-mini, and this does seem to be shared with GPT-5.4. However, the popup in the settings clearly states: You’re eligible for free daily usage on traffic shared with OpenAI. Up to 1 million tokens per…

Read more →
ChatGPT Web Update Removed Message Version Arrows (Cannot Access Edited Message History)

Support Can’t Restore 1/2 Version Arrows — Why Do Some Accounts Still Have the Version Selector (Paid Users Affected)? Paid Users Losing the Message Response Selector Bugs

version selector removed for “affected accounts” — is this intentional, and why is access unequal? retroactive removal of message version arrows/edited history — what changed, and how do users recover past variants? paid…

Read more →
Intermittent 400 "image_url is only supported by certain models" on Fine-tuned gpt-4o-2024-08-06 — Was Working Previously

Yes, it has been noted and reported. The only workaround is to be persistent and error tolerant with retries, until the issue gets so bad that the fine tuning model you paid to train is essentially useless.

It’s barely been a week of API fine-tuning gpt-4o models failing, with multiple reports and immediate replication ability. The last time the exact symptom yet complete outage happened on…

Read more →
Ben Hutchings: FOSS activity in February 2026
  • Debian packages:
    • firmware-free:
      • Bugs:
        • closed #890601: firmware-linux-free uses prebuilt blobs instead of building from source
      • Uploads:
        • uploaded version 20241210-3 to unstable
    • firmware-nonfree:
      • Bugs:
        • closed #481234: firmware-nonfree: Include firmware for p54 driver
        • closed #484177: firmware-nonfree: keyspan
          *…
Read more →
Fine-tuned 4o errors using the responses API (with image input)

The first thing I set, knowing that it would loop, was the maximum output. Useless.

This fault will incur a huge bill, unstoppable by max_output_tokens. Here, setting it to just above the minimum value accepted (which reports the same 35k input usage that would have been 301t for the small image otherwise):

Varying the ranges of max_output_tokens doesn’t vary the bill of the same amount.…

Read more →
API key permissions not working correctly

Hi and welcome back!

Have you seen this topic where a related issue has been discussed?
I believe this problem could also be caused by the UI/UX.

Missing scopes: model.request on restricted API key Bugs

This turned out to be poor UX design. If you first set “Model Capabilities” to “Request”, then unset any permissions you don’t want it works correctly. In both cases it shows “Mixed” for that…

Read more →
200 OK from /v1/responses but body times out (UND_ERR_BODY_TIMEOUT)

Although the frequency has decreased, we are still having terminated error with OpenAI Agents SDK which I believe is related to this issue.

Agents SDK -> TypeError: terminated Bugs

We use openai-agents-js (default Responses API) to run our Multi-Agent Systems (MAS). During MAS runs, we intermittently hit the following error: { "error": "TypeError: terminated" } We’ve seen this issue for a…

Read more →
Why does my vision fine-tuning job keep failing?

It is not a mere request ID, it is the whole API on both endpoints that is misconfigured and broken using image input with ft:gpt-4.1 for anyone, that has been ongoing for nearly a week.

Fine-tuned Model Vision Capabilities Lost Bugs

A fine-tuned model (ft:gpt-4.1-2025-04-14:cvm:v1-optimized2:CtvBgA1l) that previously accepted vision inputs (image_url) now rejects them with error 400. Error…

Read more →
Page 1 Older →