# \#image-reading **URL:** https://community.openai.com/tag/image-reading/266.md [Latest](https://community.openai.com/latest.md) · [Categories](https://community.openai.com/categories.md) · [Tags](https://community.openai.com/tags.md) --- ## [\[macOS\]\[ChatGPT Work\] Images unavailable to the model in cloud chats started in the desktop app](https://community.openai.com/t/macos-chatgpt-work-images-unavailable-to-the-model-in-cloud-chats-started-in-the-desktop-app/1404856)
**Author:** [@agent\_md](https://community.openai.com/u/agent_md)\ **Replies:** 0\ **Last updated:** [October 9, 2026, 8:49pm UTC](https://community.openai.com/t/macos-chatgpt-work-images-unavailable-to-the-model-in-cloud-chats-started-in-the-desktop-app/1404856 "2026-10-09T20:49:20Z")
Description Image understanding has stopped working for me in ChatGPT Work cloud chats started in the ChatGPT macOS desktop app. This began approximately 2–3 days ago, around October 6–7, 2026. Before that, it worked no… --- ## [Images in projects are not readable by ChatGPT](https://community.openai.com/t/images-in-projects-are-not-readable-by-chatgpt/1154423)
**Author:** [@Fischermann](https://community.openai.com/u/Fischermann)\ **Replies:** 6\ **Last updated:** [August 15, 2026, 4:41pm UTC](https://community.openai.com/t/images-in-projects-are-not-readable-by-chatgpt/1154423 "2026-08-15T16:41:11Z")
I uploaded images to a project in hopes that I can have the AI use it as reference. However, asking about it, ChatGPT is aware of it, the filename is correct, however it can’t seem to open it or get context from it, like… --- ## [Regarding the use of Image-2](https://community.openai.com/t/regarding-the-use-of-image-2/1379838)
**Author:** [@\_Shannon](https://community.openai.com/u/_Shannon)\ **Replies:** 5\ **Last updated:** [April 27, 2026, 4:59pm UTC](https://community.openai.com/t/regarding-the-use-of-image-2/1379838 "2026-04-27T16:59:48Z")
My current work scenario is related to the e-commerce of bras. I need to generate many e-commerce main images and detailed images about bras every day. After imgae-2 came out, I tried using image-2 to generate an e-com… --- ## [“Run failed Sorry, something went wrong.” when sending image to GPT-4o through API not even working with links](https://community.openai.com/t/run-failed-sorry-something-went-wrong-when-sending-image-to-gpt-4o-through-api-not-even-working-with-links/1119306)
**Author:** [@jack.wang](https://community.openai.com/u/jack.wang)\ **Replies:** 16\ **Last updated:** [February 14, 2026, 6:41pm UTC](https://community.openai.com/t/run-failed-sorry-something-went-wrong-when-sending-image-to-gpt-4o-through-api-not-even-working-with-links/1119306 "2026-02-14T18:41:18Z")
I am using 4o’s API to analyze some images. It used to work fine, but today it replies with: Run failed: Sorry, something went wrong. Does this happen to anyone else? --- ## [Did something change? Images from GPT Actions don’t render anymore](https://community.openai.com/t/did-something-change-images-from-gpt-actions-don-t-render-anymore/1369828)
**Author:** [@mobura](https://community.openai.com/u/mobura)\ **Replies:** 8\ **Last updated:** [January 28, 2026, 9:31am UTC](https://community.openai.com/t/did-something-change-images-from-gpt-actions-don-t-render-anymore/1369828 "2026-01-28T09:31:14Z")
Hey everyone :waving\_hand:t3: I am running into a weird issue with Custom GPT Actions and wondering if anyone else has seen this. I have an action (POST request) that returns a public PNG URL from my backend. This used… --- ## [Invalid\_image\_url error code for image urls without file extension](https://community.openai.com/t/invalid-image-url-error-code-for-image-urls-without-file-extension/1141898)
**Author:** [@warren.mcquinn](https://community.openai.com/u/warren.mcquinn)\ **Replies:** 5\ **Last updated:** [October 7, 2025, 11:49am UTC](https://community.openai.com/t/invalid-image-url-error-code-for-image-urls-without-file-extension/1141898 "2025-10-07T11:49:25Z")
At 23:25 UTC, Chat Completions API started throwing 400 timeout errors for our requests with prompt containing image\_url where the url does not have a file extension. We have images stored in S3 (CloudFront) w/o file ext… --- ## [Responses.create rewrites image URLs and returns expired SAS blobs in my MCP tool call](https://community.openai.com/t/responses-create-rewrites-image-urls-and-returns-expired-sas-blobs-in-my-mcp-tool-call/1324844)
**Author:** [@AmitosRL](https://community.openai.com/u/AmitosRL)\ **Replies:** 2\ **Last updated:** [July 28, 2025, 12:14pm UTC](https://community.openai.com/t/responses-create-rewrites-image-urls-and-returns-expired-sas-blobs-in-my-mcp-tool-call/1324844 "2025-07-28T12:14:26Z")
Hi everyone I’m building a bot that chats with users through the Assistants API. When a user attaches a photo, the bot should forward that image to our internal MCP micro-service for analysis (OCR, object detection, et… --- ## [Urgent! ChatGPT can't read file and image that uploaded from web/pc. But it's working very well when uploaded from smartphone/android app](https://community.openai.com/t/urgent-chatgpt-cant-read-file-and-image-that-uploaded-from-web-pc-but-its-working-very-well-when-uploaded-from-smartphone-android-app/1033677)
**Author:** [@peanutbutter.pakeai](https://community.openai.com/u/peanutbutter.pakeai)\ **Replies:** 203\ **Last updated:** [July 23, 2025, 7:37pm UTC](https://community.openai.com/t/urgent-chatgpt-cant-read-file-and-image-that-uploaded-from-web-pc-but-its-working-very-well-when-uploaded-from-smartphone-android-app/1033677 "2025-07-23T19:37:50Z")
When i try uploading question from image and from file through my phone, it works very well. But when i try upload from my laptop/ through chatgpt web, he says he can’t read image nor file. He says: “I can’t view or int… --- ## [Gpt4-o Support for Image URLS as tool responses](https://community.openai.com/t/gpt4-o-support-for-image-urls-as-tool-responses/907546)
**Author:** [@pabloloz](https://community.openai.com/u/pabloloz)\ **Replies:** 16\ **Last updated:** [July 19, 2025, 6:34pm UTC](https://community.openai.com/t/gpt4-o-support-for-image-urls-as-tool-responses/907546 "2025-07-19T18:34:53Z")
The previous Gpt4-turbo model supported image urls being part of a tool response which was useful for tools that responded back images to the models as a result of the tool call. The new gpt4o models have apparently los… --- ## [Upload and Analyze Multiple Files(Images, PDF) with OpenAI in Node.js](https://community.openai.com/t/upload-and-analyze-multiple-files-images-pdf-with-openai-in-node-js/1311942)
**Author:** [@Adeniji\_Adekunle\_Jam](https://community.openai.com/u/Adeniji_Adekunle_Jam)\ **Replies:** 0\ **Last updated:** [July 9, 2025, 10:31pm UTC](https://community.openai.com/t/upload-and-analyze-multiple-files-images-pdf-with-openai-in-node-js/1311942 "2025-07-09T22:31:17Z")
Have you ever needed to send multiple files (like images or PDF) to OpenAI’s API and get a meaningful response back? For example ask open ai to give you the description of the files? This tutorial shows you how to easily… --- ## [Responses API image upload error (worked before)](https://community.openai.com/t/responses-api-image-upload-error-worked-before/1289584)
**Author:** [@laimonasvailenka](https://community.openai.com/u/laimonasvailenka)\ **Replies:** 10\ **Last updated:** [July 3, 2025, 10:38pm UTC](https://community.openai.com/t/responses-api-image-upload-error-worked-before/1289584 "2025-07-03T22:38:49Z")
Hello, As of this morning, i can not pass image\_url to the responses api. API calls that have been working before (passing publick image URLS) are thrwoing this error. error message BadRequestError: Error code: 400 -… --- ## [How to upload generated image back into context using Agents SDK?](https://community.openai.com/t/how-to-upload-generated-image-back-into-context-using-agents-sdk/1291834)
**Author:** [@PromptPilot](https://community.openai.com/u/PromptPilot)\ **Replies:** 1\ **Last updated:** [June 27, 2025, 8:17am UTC](https://community.openai.com/t/how-to-upload-generated-image-back-into-context-using-agents-sdk/1291834 "2025-06-27T08:17:16Z")
Hi all, I’m working on a proof-of-concept using the OpenAI Agents SDK where an agent can generate images using the ImageGenerationTool or CodeInterpreterTool. My app stores previous messages in a database and rebuilds … --- ## [Documentation error in "Images and vision" \> "Calculating costs"](https://community.openai.com/t/documentation-error-in-images-and-vision-calculating-costs/1285072)
**Author:** [@ttpssabc](https://community.openai.com/u/ttpssabc)\ **Replies:** 1\ **Last updated:** [June 11, 2025, 5:50am UTC](https://community.openai.com/t/documentation-error-in-images-and-vision-calculating-costs/1285072 "2025-06-11T05:50:31Z")
This line should be “Scale so that the image’s shortest side is less than 768px” --- ## [Thumbnail versions of pictures used in ChatGPT](https://community.openai.com/t/thumbnail-versions-of-pictures-used-in-chatgpt/1284983)
**Author:** [@ronaldanthony4](https://community.openai.com/u/ronaldanthony4)\ **Replies:** 0\ **Last updated:** [June 11, 2025, 1:23am UTC](https://community.openai.com/t/thumbnail-versions-of-pictures-used-in-chatgpt/1284983 "2025-06-11T01:23:49Z")
In one chat, I generated so many pictures and also uploaded many pictures in one chat, but as many picture goes by, the chat became laggy, maybe because it takes too much time to show all its generated pictures in HD. So… --- ## [Realtime API Images as input](https://community.openai.com/t/realtime-api-images-as-input/1089108)
**Author:** [@gmudcxr](https://community.openai.com/u/gmudcxr)\ **Replies:** 0\ **Last updated:** [January 13, 2025, 2:35am UTC](https://community.openai.com/t/realtime-api-images-as-input/1089108 "2025-01-13T02:35:07Z")
Hello, I am trying to use the Realtime API (WebSocket) to analyze some images. I assume (hope) the main issue is syntax, as documentation seems limited. I am trying to send each message as follows: event = { “type”: “… --- ## [Image files as context for Agents](https://community.openai.com/t/image-files-as-context-for-agents/1246935)
**Author:** [@dexter.mancuso](https://community.openai.com/u/dexter.mancuso)\ **Replies:** 0\ **Last updated:** [April 30, 2025, 1:13am UTC](https://community.openai.com/t/image-files-as-context-for-agents/1246935 "2025-04-30T01:13:55Z")
I’m looking at the context management section of the docs and trying to figure out how I could have multiple image files as local context for an agent. For example, I could have an image of a car, of a train, plane and b… --- ## [Gpt 4o sometime not able to access image from url](https://community.openai.com/t/gpt-4o-sometime-not-able-to-access-image-from-url/903528)
**Author:** [@ashrafkhan0997](https://community.openai.com/u/ashrafkhan0997)\ **Replies:** 3\ **Last updated:** [April 7, 2025, 5:38pm UTC](https://community.openai.com/t/gpt-4o-sometime-not-able-to-access-image-from-url/903528 "2025-04-07T17:38:04Z")
I am sending an image url to the gpt4o(azure openai endpoint). The model is throwing an error openai.BadRequestError: Error code: 400 - {'error': {'code': 'BadRequest', 'message': 'The provided image url is not accessi… --- ## [Maximum number of images for Responses & Assistants API?](https://community.openai.com/t/maximum-number-of-images-for-responses-assistants-api/1148105)
**Author:** [@ianwatts](https://community.openai.com/u/ianwatts)\ **Replies:** 1\ **Last updated:** [March 25, 2025, 11:40pm UTC](https://community.openai.com/t/maximum-number-of-images-for-responses-assistants-api/1148105 "2025-03-25T23:40:08Z")
Is there any official word on what the maximum number of images you can send to the Responses or Assistants API is (specifically for gpt-4o with detail-low)? In the past I’ve been able to send up to 250 images to the Ass… --- ## [Prompt for Image to JSON conversion](https://community.openai.com/t/prompt-for-image-to-json-conversion/866883)
**Author:** [@cg27031998](https://community.openai.com/u/cg27031998)\ **Replies:** 3\ **Last updated:** [March 15, 2025, 1:17pm UTC](https://community.openai.com/t/prompt-for-image-to-json-conversion/866883 "2025-03-15T13:17:57Z")
I have process diagram present in a image. I am converting this image to base64 encoding and attaching it to my prompt. I have given instructions to prompt to understand the image and give output in the JSON structure wh… --- ## [Assistant API can not handle image URLs without extension](https://community.openai.com/t/assistant-api-can-not-handle-image-urls-without-extension/929717)
**Author:** [@kvk1986](https://community.openai.com/u/kvk1986)\ **Replies:** 3\ **Last updated:** [March 13, 2025, 1:37am UTC](https://community.openai.com/t/assistant-api-can-not-handle-image-urls-without-extension/929717 "2025-03-13T01:37:54Z")
While the Chat Completions API can handle image URLs without extensions, the Assistants API currently requires the URL to end with an extension. I would like to submit this as a Feature Request as this discrepancy is qu… --- ## [GTP-4o no longer analyses images](https://community.openai.com/t/gtp-4o-no-longer-analyses-images/982681)
**Author:** [@360precision](https://community.openai.com/u/360precision)\ **Replies:** 2\ **Last updated:** [October 17, 2024, 2:40pm UTC](https://community.openai.com/t/gtp-4o-no-longer-analyses-images/982681 "2024-10-17T14:40:31Z")
I know there are other similar topics but maybe if this issue becomes more obvious it might get solve, I’ve been using gtp-4o to analyse images via the API to return a desciption in JSON format. The code has been workin… --- ## [Using GPT-4-Turbo to fill out complex PDF forms](https://community.openai.com/t/using-gpt-4-turbo-to-fill-out-complex-pdf-forms/722020)
**Author:** [@kevob33](https://community.openai.com/u/kevob33)\ **Replies:** 6\ **Last updated:** [February 28, 2025, 4:06pm UTC](https://community.openai.com/t/using-gpt-4-turbo-to-fill-out-complex-pdf-forms/722020 "2025-02-28T16:06:48Z")
Without providing many details, my company is an intermediary between other companies (clients) and end users (customers). We receive many different PDF forms from these clients which must be filled in with many customer… --- ## [Summarization call on image sometimes returns empty string today after working for weeks](https://community.openai.com/t/summarization-call-on-image-sometimes-returns-empty-string-today-after-working-for-weeks/1114567)
**Author:** [@fasttosmile](https://community.openai.com/u/fasttosmile)\ **Replies:** 1\ **Last updated:** [February 8, 2025, 3:17am UTC](https://community.openai.com/t/summarization-call-on-image-sometimes-returns-empty-string-today-after-working-for-weeks/1114567 "2025-02-08T03:17:27Z")
Has there been some change to how images are supposed to be processed for summarization? I have an image that I’ve been using for over a month (processing repeatedly to test smth else), and yesterday the 4o model summar… --- ## [Cannot pass an image to GPT-4o](https://community.openai.com/t/cannot-pass-an-image-to-gpt-4o/1098895)
**Author:** [@kevin.guimard](https://community.openai.com/u/kevin.guimard)\ **Replies:** 2\ **Last updated:** [January 23, 2025, 9:39am UTC](https://community.openai.com/t/cannot-pass-an-image-to-gpt-4o/1098895 "2025-01-23T09:39:47Z")
I have a valid PIL image (size 1654 x 2339) that I try to include in my query, to no avail. The code is the following (with langchain): buffer = BytesIO() image.save(buffer, format=“JPEG”) img\_str = base64.b64encode… --- ## [Batching images to any GPT models](https://community.openai.com/t/batching-images-to-any-gpt-models/946601)
**Author:** [@dev94](https://community.openai.com/u/dev94)\ **Replies:** 3\ **Last updated:** [January 21, 2025, 12:17pm UTC](https://community.openai.com/t/batching-images-to-any-gpt-models/946601 "2025-01-21T12:17:15Z")
I’m trying to add batching for a function that I have. The function sends images to the api using base64. The error I was encountering is that the batch has a parameter that is asking for urls which I’m trying to avoid i… --- ## [Serious bugs in ChatGPT for Windows and Web browser: Why aren't the new models working?](https://community.openai.com/t/serious-bugs-in-chatgpt-for-windows-and-web-browser-why-arent-the-new-models-working/1040428)
**Author:** [@AI82](https://community.openai.com/u/AI82)\ **Replies:** 1\ **Last updated:** [January 20, 2025, 4:51pm UTC](https://community.openai.com/t/serious-bugs-in-chatgpt-for-windows-and-web-browser-why-arent-the-new-models-working/1040428 "2025-01-20T16:51:23Z")
For the past few days, I’ve been encountering an issue where, whether I’m using a web browser or the ChatGPT app from the Microsoft Store on Windows (which is essentially a web browser wrapped in a more convenient interf… --- ## [Identify meeting-room tables and annotate images](https://community.openai.com/t/identify-meeting-room-tables-and-annotate-images/1096531)
**Author:** [@stabenfeldt](https://community.openai.com/u/stabenfeldt)\ **Replies:** 3\ **Last updated:** [January 20, 2025, 3:17pm UTC](https://community.openai.com/t/identify-meeting-room-tables-and-annotate-images/1096531 "2025-01-20T15:17:37Z")
Hi, I’ve been working as a developer for many years, but I’m new to using tools like OpenAI. I’m building a meeting room configurator. I want the user journey to be as follows: A user uploads an image of a floor pla… --- ## [How does the image OCR actually work in gpt-4?](https://community.openai.com/t/how-does-the-image-ocr-actually-work-in-gpt-4/785890)
**Author:** [@steveny](https://community.openai.com/u/steveny)\ **Replies:** 2\ **Last updated:** [January 9, 2025, 3:19pm UTC](https://community.openai.com/t/how-does-the-image-ocr-actually-work-in-gpt-4/785890 "2025-01-09T15:19:53Z")
I asked gpt-4 to process my receipt image and return in a json format. It does work pretty accurately. Moreover, it provides the View analysis which includes Python code that uses pytesseract to generate the result. Howe… --- ## [Token count difference between image url and base64 for same image in api call](https://community.openai.com/t/token-count-difference-between-image-url-and-base64-for-same-image-in-api-call/857754)
**Author:** [@AA77](https://community.openai.com/u/AA77)\ **Replies:** 1\ **Last updated:** [January 6, 2025, 7:25pm UTC](https://community.openai.com/t/token-count-difference-between-image-url-and-base64-for-same-image-in-api-call/857754 "2025-01-06T19:25:49Z")
If I make an api call for same image but one with url and another with base64, will there be difference between token count and cost? --- ## [How do I edit an image through the prompt](https://community.openai.com/t/how-do-i-edit-an-image-through-the-prompt/712949)
**Author:** [@fernandoimmersive](https://community.openai.com/u/fernandoimmersive)\ **Replies:** 10\ **Last updated:** [December 31, 2024, 3:27pm UTC](https://community.openai.com/t/how-do-i-edit-an-image-through-the-prompt/712949 "2024-12-31T15:27:40Z")
hello, how do I edit an image through the prompt without using mask. I’ve tried several ways. I used this way: curl -H 'Authorization: Bearer sk-\*\*\*uPb' https://api.openai.com/v1/images/edits --form 'image=@"./tmp/sel… [Next page](https://community.openai.com/tag/image-reading/266.md?match_all_tags=true&page=1&tags%5B%5D=image-reading)