# \#gpt-4-vision **URL:** https://community.openai.com/tag/gpt-4-vision/312.md [Latest](https://community.openai.com/latest.md) · [Categories](https://community.openai.com/categories.md) · [Tags](https://community.openai.com/tags.md) --- ## [Getting GPT Vision To Return Coordinates](https://community.openai.com/t/getting-gpt-vision-to-return-coordinates/671669)
**Author:** [@dano1066](https://community.openai.com/u/dano1066)\ **Replies:** 12\ **Last updated:** [September 19, 2026, 6:33pm UTC](https://community.openai.com/t/getting-gpt-vision-to-return-coordinates/671669 "2026-09-19T18:33:03Z")
Vision has been doing a really good job at describing the images i provide but i want to try and also make it give me back the x/y coordinates of the specific things in the image so that i can use javascript to create a … --- ## [Video analysis with Open AI](https://community.openai.com/t/video-analysis-with-open-ai/852531)
**Author:** [@lreinhard7](https://community.openai.com/u/lreinhard7)\ **Replies:** 4\ **Last updated:** [March 31, 2026, 9:44am UTC](https://community.openai.com/t/video-analysis-with-open-ai/852531 "2026-03-31T09:44:04Z")
Hi, I am trying to analyze a video that I made with my drone. The video has no voice but rather records activities only. What I try to get is a summary of what happened in the video. Example: 08:01 Person leaves hou… --- ## [“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? --- ## [Discrepancy: omni-moderation-latest Token Usage vs Tier 1 Rate Limits (Vision)](https://community.openai.com/t/discrepancy-omni-moderation-latest-token-usage-vs-tier-1-rate-limits-vision/1371064)
**Author:** [@Anass\_El\_Basraoui](https://community.openai.com/u/Anass_El_Basraoui)\ **Replies:** 4\ **Last updated:** [January 7, 2026, 1:40pm UTC](https://community.openai.com/t/discrepancy-omni-moderation-latest-token-usage-vs-tier-1-rate-limits-vision/1371064 "2026-01-07T13:40:43Z")
Hello everyone, I am implementing an image moderation pipeline using the \`omni-moderation-latest\` model. I am currently on Tier 1, which states a limit of 10,000 TPM (Tokens Per Minute)→ https://platform.openai.com/docs… --- ## [Is it possible to do visual comparison with OpenAI Vision?](https://community.openai.com/t/is-it-possible-to-do-visual-comparison-with-openai-vision/1271005)
**Author:** [@Sanjeev\_Bhandari](https://community.openai.com/u/Sanjeev_Bhandari)\ **Replies:** 2\ **Last updated:** [December 28, 2025, 4:19am UTC](https://community.openai.com/t/is-it-possible-to-do-visual-comparison-with-openai-vision/1271005 "2025-12-28T04:19:47Z")
Hi there, I’m interested in comparing two images to analyze layout differences—specifically, comparing a Figma design with the corresponding webpage layout. Is it possible to perform this kind of visual comparison usin… --- ## [Responses API: STILL overbilling on image inputs (BUG/ISSUE)](https://community.openai.com/t/responses-api-still-overbilling-on-image-inputs-bug-issue/1368505)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 5\ **Last updated:** [December 17, 2025, 9:28am UTC](https://community.openai.com/t/responses-api-still-overbilling-on-image-inputs-bug-issue/1368505 "2025-12-17T09:28:03Z")
Issue: detail setting is not being respected on the Responses API for the majority of models. I’ve just run through ALL the “vision” API models that accept detail: low. While Chat Completions is correct, the Responses A… --- ## [Chat Completions vs Responses and pdf file (new PDF file vision upload modality added to CC)](https://community.openai.com/t/chat-completions-vs-responses-and-pdf-file-new-pdf-file-vision-upload-modality-added-to-cc/1143115)
**Author:** [@lszczygielek](https://community.openai.com/u/lszczygielek)\ **Replies:** 9\ **Last updated:** [December 6, 2025, 4:47pm UTC](https://community.openai.com/t/chat-completions-vs-responses-and-pdf-file-new-pdf-file-vision-upload-modality-added-to-cc/1143115 "2025-12-06T16:47:48Z")
I recently started testing the OpenAI GPT API and came across a strange situation while experimenting in the Playground. When I send a file for summarization, the Chat Completions API consumes significantly fewer input … --- ## [Unable to upload big JSONL for Vision Model Fine Tuning](https://community.openai.com/t/unable-to-upload-big-jsonl-for-vision-model-fine-tuning/1367745)
**Author:** [@abhishek.ss](https://community.openai.com/u/abhishek.ss)\ **Replies:** 1\ **Last updated:** [November 28, 2025, 11:35am UTC](https://community.openai.com/t/unable-to-upload-big-jsonl-for-vision-model-fine-tuning/1367745 "2025-11-28T11:35:45Z")
I am trying to fine tune gpt-4o-2024-08-06 following the guidelines at: https://platform.openai.com/docs/guides/vision-fine-tuning I have created JSONL file for training and validation with embedded base64 encoded imag… --- ## [Document transcription & metadata extraction Project - vision models - Best Practices](https://community.openai.com/t/document-transcription-metadata-extraction-project-vision-models-best-practices/1359981)
**Author:** [@hernanverzini](https://community.openai.com/u/hernanverzini)\ **Replies:** 1\ **Last updated:** [September 26, 2025, 8:57pm UTC](https://community.openai.com/t/document-transcription-metadata-extraction-project-vision-models-best-practices/1359981 "2025-09-26T20:57:49Z")
Hello everyone, I’m currently evaluating a project that requires transcription and data extraction from more than 1,000,000 pagesusing OpenAI vision models. I’d love to get some advice from the community on three key p… --- ## [Stop the overbilling of input images to gpt-4o-mini!](https://community.openai.com/t/stop-the-overbilling-of-input-images-to-gpt-4o-mini/1333507)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 0\ **Last updated:** [August 4, 2025, 12:41pm UTC](https://community.openai.com/t/stop-the-overbilling-of-input-images-to-gpt-4o-mini/1333507 "2025-08-04T12:41:08Z")
It’s been a year of inflated prices sending input images to gpt-4o-mini for vision, and needing to tell people not to use this model, because it actually costs more per image than the full gpt-4o model. The price inflat… --- ## [How do I calculate image tokens in GPT4 Vision?](https://community.openai.com/t/how-do-i-calculate-image-tokens-in-gpt4-vision/492318)
**Author:** [@niyazvakhpiev1](https://community.openai.com/u/niyazvakhpiev1)\ **Replies:** 13\ **Last updated:** [July 24, 2025, 2:12pm UTC](https://community.openai.com/t/how-do-i-calculate-image-tokens-in-gpt4-vision/492318 "2025-07-24T14:12:57Z")
Hi, how do I count how many number of tokens does the each image has when using gpt-4-vision-preview model? --- ## [GPT-4-Vision Error - BadRequestError: 400 Invalid image](https://community.openai.com/t/gpt-4-vision-error-badrequesterror-400-invalid-image/516486)
**Author:** [@georg-san](https://community.openai.com/u/georg-san)\ **Replies:** 8\ **Last updated:** [July 23, 2025, 11:45am UTC](https://community.openai.com/t/gpt-4-vision-error-badrequesterror-400-invalid-image/516486 "2025-07-23T11:45:22Z")
Using gpt-4-vision-preview I’m getting a BadRequestError: 400 Invalid image error when the image url has some encoding. Specifically, I’m using a getDownloadUrl() from Firebase storage which has some characters encoded … --- ## [Returning image as tool output in Assistants API?](https://community.openai.com/t/returning-image-as-tool-output-in-assistants-api/763387)
**Author:** [@turbolucius](https://community.openai.com/u/turbolucius)\ **Replies:** 4\ **Last updated:** [June 4, 2025, 4:45pm UTC](https://community.openai.com/t/returning-image-as-tool-output-in-assistants-api/763387 "2025-06-04T16:45:45Z")
Hello! I’m trying to figure out if there’s a way to return an image as a tool output for Assistants. I can think of several use cases where that would be useful (I’m making a small app that can access a “take screenshot… --- ## [How to get gpt-vision-preview in api call to correctly read a publicly accessible jpg image from Google Drive?](https://community.openai.com/t/how-to-get-gpt-vision-preview-in-api-call-to-correctly-read-a-publicly-accessible-jpg-image-from-google-drive/659490)
**Author:** [@geoffreyofbordentown](https://community.openai.com/u/geoffreyofbordentown)\ **Replies:** 4\ **Last updated:** [May 29, 2025, 2:13am UTC](https://community.openai.com/t/how-to-get-gpt-vision-preview-in-api-call-to-correctly-read-a-publicly-accessible-jpg-image-from-google-drive/659490 "2025-05-29T02:13:51Z")
Error message: openai.BadRequestError: Error code: 400 - {‘error’: {‘message’: “You uploaded an unsupported image. Please make sure your image is below 20 MB in size and is of one the following formats: \[‘png’, ‘jpeg’, … --- ## [Vision: More than one way to send a request, undocumented: and different pricing realized](https://community.openai.com/t/vision-more-than-one-way-to-send-a-request-undocumented-and-different-pricing-realized/1262691)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 2\ **Last updated:** [May 16, 2025, 6:30pm UTC](https://community.openai.com/t/vision-more-than-one-way-to-send-a-request-undocumented-and-different-pricing-realized/1262691 "2025-05-16T18:30:21Z")
First plug: a web page to make vision pricing clearer. Add an image, by either upload or URL, or simply simulating the dimensions. Add multiple images; Select the (practical) model that you want to see the price calc… --- ## [Using image urls in the chat completion endpoint for vision](https://community.openai.com/t/using-image-urls-in-the-chat-completion-endpoint-for-vision/965024)
**Author:** [@tanm1](https://community.openai.com/u/tanm1)\ **Replies:** 5\ **Last updated:** [May 8, 2025, 1:48am UTC](https://community.openai.com/t/using-image-urls-in-the-chat-completion-endpoint-for-vision/965024 "2025-05-08T01:48:43Z")
I am using image urls as parameters in the chat completion endpoint. The user will input their image and then that gets uploaded to a storage bucket and a url for that image is created. The image url is then used as a pa… --- ## [ChatGPT feature request: allow removal of input images from chat sessions](https://community.openai.com/t/chatgpt-feature-request-allow-removal-of-input-images-from-chat-sessions/1191674)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 2\ **Last updated:** [January 2, 2026, 10:33am UTC](https://community.openai.com/t/chatgpt-feature-request-allow-removal-of-input-images-from-chat-sessions/1191674 "2026-01-02T10:33:28Z")
No AI written form letter here. Just a simple request (for likely a good bit of work to power it): Let an image be deleted from any or all prior ChatGPT messages. When finally submitted, this could result in a “branch… --- ## [Does OpenAI prohibit OCR?](https://community.openai.com/t/does-openai-prohibit-ocr/1232508)
**Author:** [@dandyfiner](https://community.openai.com/u/dandyfiner)\ **Replies:** 5\ **Last updated:** [April 25, 2025, 4:08pm UTC](https://community.openai.com/t/does-openai-prohibit-ocr/1232508 "2025-04-25T16:08:04Z")
The documentation says input images may not contain text. Specifically, under Image Input Requirements → Other Requirements it says, “No Text”. I just want to verify: does this means using OpenAI APIs for OCR is prohi… --- ## [GPT-4.1 vision price calculations -- incorrect billing on full model](https://community.openai.com/t/gpt-4-1-vision-price-calculations-incorrect-billing-on-full-model/1230130)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 7\ **Last updated:** [April 24, 2025, 3:22pm UTC](https://community.openai.com/t/gpt-4-1-vision-price-calculations-incorrect-billing-on-full-model/1230130 "2025-04-24T15:22:06Z")
Billing issue with image tokens on API Playground for gpt-4.1 Regardless of the input image size sent, I get the same input tokens GPT 4.1 image billing: blank 1800x2400: 773t up content 1800x2400 gif: 773t content: 240… --- ## [Has GPT-4 Vision Preview been replaced or removed? Getting \[model\_not\_found\] error](https://community.openai.com/t/has-gpt-4-vision-preview-been-replaced-or-removed-getting-model-not-found-error/1151485)
**Author:** [@seco](https://community.openai.com/u/seco)\ **Replies:** 2\ **Last updated:** [April 20, 2025, 6:22am UTC](https://community.openai.com/t/has-gpt-4-vision-preview-been-replaced-or-removed-getting-model-not-found-error/1151485 "2025-04-20T06:22:51Z")
Hello everyone, I’ve been developing an iOS app that used to rely on the gpt-4-vision-preview model to generate text from product images. However, I’m now getting the following error: csharp Copiar \[model\_not\_found\] … --- ## [GPT-4-turbo vision API recognizes image\_url as base64 encoded image data](https://community.openai.com/t/gpt-4-turbo-vision-api-recognizes-image-url-as-base64-encoded-image-data/734243)
**Author:** [@ekgnlqlem](https://community.openai.com/u/ekgnlqlem)\ **Replies:** 10\ **Last updated:** [April 2, 2025, 9:22am UTC](https://community.openai.com/t/gpt-4-turbo-vision-api-recognizes-image-url-as-base64-encoded-image-data/734243 "2025-04-02T09:22:20Z")
Provided is the section of my typescript code. s3ImageURL and resolution variables have valid values. s3ImageURL looks like https://s3.ap-northeast-2.amazonaws.com/bucketName/fileName resolution - “auto” | “low” | “hig… --- ## [GPT-4-o-Mini Vision Token Cost Issue](https://community.openai.com/t/gpt-4-o-mini-vision-token-cost-issue/989143)
**Author:** [@jordan-coursey](https://community.openai.com/u/jordan-coursey)\ **Replies:** 2\ **Last updated:** [March 26, 2025, 6:42am UTC](https://community.openai.com/t/gpt-4-o-mini-vision-token-cost-issue/989143 "2025-03-26T06:42:56Z")
Hi everyone, I wanted to bring attention to what appears to be a significant token cost discrepancy with GPT-4-o-Mini’s image handling. I’ve noticed that when processing the exact same images with identical code, simply… --- ## [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… --- ## [Improving GPT Vision for multi-column document analysis](https://community.openai.com/t/improving-gpt-vision-for-multi-column-document-analysis/1151453)
**Author:** [@akunbaru](https://community.openai.com/u/akunbaru)\ **Replies:** 0\ **Last updated:** [March 25, 2025, 5:44am UTC](https://community.openai.com/t/improving-gpt-vision-for-multi-column-document-analysis/1151453 "2025-03-25T05:44:20Z")
Hi everyone, I’m currently developing a contract analysis tool called, which uses GPT-4 Vision (GPT-4o) to extract structured information from PDF or DOCX contract documents. The system works by processing each page visu… --- ## [API ISSUE: "Responses" endpoint: using vision with user image is only errors (now fixed)](https://community.openai.com/t/api-issue-responses-endpoint-using-vision-with-user-image-is-only-errors-now-fixed/1142676)
**Author:** [@\_j](https://community.openai.com/u/_j)\ **Replies:** 4\ **Last updated:** [March 15, 2025, 3:20am UTC](https://community.openai.com/t/api-issue-responses-endpoint-using-vision-with-user-image-is-only-errors-now-fixed/1142676 "2025-03-15T03:20:03Z")
In a recent task, I had to switch from Responses to Chat Completions in the playground (which works on the same context). A user input message with an image, or “additional messages” with an image added (depicted), r… --- ## [GPT-4-vision extraction of tables with branched rows/vertically-merged cells](https://community.openai.com/t/gpt-4-vision-extraction-of-tables-with-branched-rows-vertically-merged-cells/728749)
**Author:** [@Azure231](https://community.openai.com/u/Azure231)\ **Replies:** 9\ **Last updated:** [March 8, 2025, 2:12pm UTC](https://community.openai.com/t/gpt-4-vision-extraction-of-tables-with-branched-rows-vertically-merged-cells/728749 "2025-03-08T14:12:42Z")
I’m struggling with a specific use-case and wondering if anyone has any insight. I have been using GPT-4-vision to extract data from medical test documents. There are a range of different formats. GPT-4-vision does ve… --- ## [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… --- ## [Serializing image as it was sent by the AI](https://community.openai.com/t/serializing-image-as-it-was-sent-by-the-ai/1130619)
**Author:** [@thiago.wander22](https://community.openai.com/u/thiago.wander22)\ **Replies:** 1\ **Last updated:** [February 27, 2025, 2:12am UTC](https://community.openai.com/t/serializing-image-as-it-was-sent-by-the-ai/1130619 "2025-02-27T02:12:24Z")
I have an application that mixes AI and python code for sending messages to the user. The model can choose to use a tool send\_image to the user, and the tool will send an image to the user. Right now, when the user inte… --- ## [Can't seem to eliminate markdown format](https://community.openai.com/t/cant-seem-to-eliminate-markdown-format/723260)
**Author:** [@fredrikjohansen01](https://community.openai.com/u/fredrikjohansen01)\ **Replies:** 15\ **Last updated:** [February 21, 2025, 2:51pm UTC](https://community.openai.com/t/cant-seem-to-eliminate-markdown-format/723260 "2025-02-21T14:51:58Z")
I am trying very hard to avoid Markdown format Example In my responses, as I want my response to be HTML only. “messages”: \[ { “role”: “system”, “content”: \[ { “type”: “text”, “text”: “You are a 1-response API. Ma… --- ## [Best practice scanned PDF / What model to use?](https://community.openai.com/t/best-practice-scanned-pdf-what-model-to-use/1122717)
**Author:** [@JorenDeb](https://community.openai.com/u/JorenDeb)\ **Replies:** 3\ **Last updated:** [February 19, 2025, 6:45am UTC](https://community.openai.com/t/best-practice-scanned-pdf-what-model-to-use/1122717 "2025-02-19T06:45:30Z")
Hi all I’m new to the OpenAI API. I’ve written a (backoffice) application which uploads documents (mainly pdf) to OpenAI to extract data. All works perfectly, but i’m struggling with scanned pdf’s. What the best practi… [Next page](https://community.openai.com/tag/gpt-4-vision/312.md?match_all_tags=true&page=1&tags%5B%5D=gpt-4-vision)