# \#gpt4-vision **URL:** https://community.openai.com/tag/gpt4-vision/323.md [Latest](https://community.openai.com/latest.md) · [Categories](https://community.openai.com/categories.md) · [Tags](https://community.openai.com/tags.md) --- ## [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? --- ## [Streaming support in Azure Openai GPT 4 Vision model](https://community.openai.com/t/streaming-support-in-azure-openai-gpt-4-vision-model/1229127)
**Author:** [@svelidanda](https://community.openai.com/u/svelidanda)\ **Replies:** 0\ **Last updated:** [April 10, 2025, 9:02am UTC](https://community.openai.com/t/streaming-support-in-azure-openai-gpt-4-vision-model/1229127 "2025-04-10T09:02:10Z")
Is streaming supported in Azure Openai GPT 4 Vision model? If supported, how can we use stream\_options? --- ## [Summarizing process flow/ block diagrams](https://community.openai.com/t/summarizing-process-flow-block-diagrams/1137576)
**Author:** [@kshtjvishnoi](https://community.openai.com/u/kshtjvishnoi)\ **Replies:** 1\ **Last updated:** [March 7, 2025, 12:00pm UTC](https://community.openai.com/t/summarizing-process-flow-block-diagrams/1137576 "2025-03-07T12:00:35Z")
I have been testing gpt4-vison to analyze and summarize block/process flow diagrams. I see that text extraction is good but I am facing issues with arrows, especially the ones going right to left. Any suggestions on how … --- ## [Vision Prompting Does not seem to be accurate](https://community.openai.com/t/vision-prompting-does-not-seem-to-be-accurate/1126552)
**Author:** [@promptjudy](https://community.openai.com/u/promptjudy)\ **Replies:** 1\ **Last updated:** [February 21, 2025, 11:29pm UTC](https://community.openai.com/t/vision-prompting-does-not-seem-to-be-accurate/1126552 "2025-02-21T23:29:02Z")
Have people tried using vision models to perform PDF rag? What is the type of accuracy you are seeing? Even the latest models arent able to quite read pdf documents without actual text provided (OCR) - or is this a promp… --- ## [How to calculate input and output tokens?](https://community.openai.com/t/how-to-calculate-input-and-output-tokens/1026209)
**Author:** [@haseeb21](https://community.openai.com/u/haseeb21)\ **Replies:** 0\ **Last updated:** [November 21, 2024, 5:06am UTC](https://community.openai.com/t/how-to-calculate-input-and-output-tokens/1026209 "2024-11-21T05:06:46Z")
I am using Vision Capability of gpt4o using API key, I want to know what is the total tokens of input, the image I am giving, and the output that is generated. Following is the code: import base64 from openai import Ope… --- ## [ChatGPT 4 Vision sometimes returns "I'm sorry, I can't provide help with that request" and sometimes actual response for the same Image](https://community.openai.com/t/chatgpt-4-vision-sometimes-returns-im-sorry-i-cant-provide-help-with-that-request-and-sometimes-actual-response-for-the-same-image/594761)
**Author:** [@anon83257046](https://community.openai.com/u/anon83257046)\ **Replies:** 21\ **Last updated:** [October 19, 2024, 5:51pm UTC](https://community.openai.com/t/chatgpt-4-vision-sometimes-returns-im-sorry-i-cant-provide-help-with-that-request-and-sometimes-actual-response-for-the-same-image/594761 "2024-10-19T17:51:06Z")
I’m having a weird issue with ChatGPT 4 Vision API. I’m running this request (which mostly works): let payload: \[String: Any\] = \[ "model": "gpt-4-vision-preview", "messages": \[ \[ … --- ## [Gpt4-vision refuses to create dalle prompt](https://community.openai.com/t/gpt4-vision-refuses-to-create-dalle-prompt/493762)
**Author:** [@jack.richard.long](https://community.openai.com/u/jack.richard.long)\ **Replies:** 4\ **Last updated:** [September 12, 2024, 6:21am UTC](https://community.openai.com/t/gpt4-vision-refuses-to-create-dalle-prompt/493762 "2024-09-12T06:21:06Z")
GPT4 Vision refuses to create a prompt for dalle. Prompt: “This image is my logo. Create a dalle prompt to generate a improved version. Please respond only with the prompt.”; Image: b64 encoded About 1/3 the time I ge… --- ## [Increased Error Rate on GPT-4o Vision](https://community.openai.com/t/increased-error-rate-on-gpt-4o-vision/867921)
**Author:** [@georg-san](https://community.openai.com/u/georg-san)\ **Replies:** 5\ **Last updated:** [August 5, 2024, 5:09pm UTC](https://community.openai.com/t/increased-error-rate-on-gpt-4o-vision/867921 "2024-08-05T17:09:39Z")
We’re seeing an increased error rate with GPT-4o vision. Est. ~5-10% of requests are failing (in production!). This started to spike in the last 3h. The error returned from the completion: BadRequestError: 400 You uplo… --- ## [Tried everything with RateLimitError: Error code: 429 with gpt4-o](https://community.openai.com/t/tried-everything-with-ratelimiterror-error-code-429-with-gpt4-o/780065)
**Author:** [@ruki](https://community.openai.com/u/ruki)\ **Replies:** 3\ **Last updated:** [August 3, 2024, 3:37pm UTC](https://community.openai.com/t/tried-everything-with-ratelimiterror-error-code-429-with-gpt4-o/780065 "2024-08-03T15:37:17Z")
Hi, I have frustrating issue. Let me give the code to save all of our time. I’ve seen the similar questions, but their solutions didn’t work. I already pass the base64 images using image\_url as suggested in those posts… --- ## [The performance difference between ChatGPT4o and gpt4o api using the same prompt for image analysis](https://community.openai.com/t/the-performance-difference-between-chatgpt4o-and-gpt4o-api-using-the-same-prompt-for-image-analysis/821401)
**Author:** [@HuixinZhong](https://community.openai.com/u/HuixinZhong)\ **Replies:** 5\ **Last updated:** [July 27, 2024, 6:50am UTC](https://community.openai.com/t/the-performance-difference-between-chatgpt4o-and-gpt4o-api-using-the-same-prompt-for-image-analysis/821401 "2024-07-27T06:50:02Z")
When I use the same prompt shown as follows to analyze images using chatgpt4o and gpt4o api. Why I get very different results: very good result from chatgot4o while totally wrong answers from gpt4o api? wondering why … --- ## [Worse OCR on rotated Text](https://community.openai.com/t/worse-ocr-on-rotated-text/461827)
**Author:** [@misch221](https://community.openai.com/u/misch221)\ **Replies:** 8\ **Last updated:** [June 26, 2024, 8:18pm UTC](https://community.openai.com/t/worse-ocr-on-rotated-text/461827 "2024-06-26T20:18:03Z")
So I’m using the image input for reading labels off of beverage cans and bottles. What I noticed is that it performs very well if the text is in the right rotation(way better than all OCR-tools I have tested so far). But… --- ## [I get the error "Run failed Sorry, something went wrong." when trying to send images](https://community.openai.com/t/i-get-the-error-run-failed-sorry-something-went-wrong-when-trying-to-send-images/839215)
**Author:** [@vinicius.ruffo](https://community.openai.com/u/vinicius.ruffo)\ **Replies:** 1\ **Last updated:** [June 26, 2024, 12:48am UTC](https://community.openai.com/t/i-get-the-error-run-failed-sorry-something-went-wrong-when-trying-to-send-images/839215 "2024-06-26T00:48:54Z")
Today I was testing an assistant and I saw that when sending an image either via API or interface I received the error “Run failed Sorry, something went wrong.” --- ## [GPT-4o Vision help Image input not working](https://community.openai.com/t/gpt-4o-vision-help-image-input-not-working/824317)
**Author:** [@aarushnangunoori](https://community.openai.com/u/aarushnangunoori)\ **Replies:** 1\ **Last updated:** [June 17, 2024, 2:50pm UTC](https://community.openai.com/t/gpt-4o-vision-help-image-input-not-working/824317 "2024-06-17T14:50:43Z")
I need my program to take a screenshot of the screen and then upload that into gpt4(I would like to use gpt4o if possible) , search the internet and then give me a response based off what its sees. But I cant seem to fig… --- ## [Analyzing images that may be NSFW using gpt-4-vision](https://community.openai.com/t/analyzing-images-that-may-be-nsfw-using-gpt-4-vision/788289)
**Author:** [@RenderHub](https://community.openai.com/u/RenderHub)\ **Replies:** 0\ **Last updated:** [May 30, 2024, 9:45pm UTC](https://community.openai.com/t/analyzing-images-that-may-be-nsfw-using-gpt-4-vision/788289 "2024-05-30T21:45:33Z")
So, I know generally speaking gpt cant be used to generate NSFW images, but in our case we are sort of doing the reverse. We are generating keywords from existing images. These images are user submitted and it would be n… --- ## [Limitations of GPT-4V's high res tiling process?](https://community.openai.com/t/limitations-of-gpt-4vs-high-res-tiling-process/712179)
**Author:** [@mwp897](https://community.openai.com/u/mwp897)\ **Replies:** 1\ **Last updated:** [April 9, 2024, 11:07pm UTC](https://community.openai.com/t/limitations-of-gpt-4vs-high-res-tiling-process/712179 "2024-04-09T23:07:11Z")
In using GPT-4V for tables that are significantly bigger than 512x512 in high res mode, I’m finding pretty frequent row and column confusion, where the right value is extracted in the wrong row/column. However when I con… --- ## [Cost of Vision using GPT-4o](https://community.openai.com/t/cost-of-vision-using-gpt-4o/775002)
**Author:** [@tmrgn](https://community.openai.com/u/tmrgn)\ **Replies:** 1\ **Last updated:** [May 27, 2024, 2:04pm UTC](https://community.openai.com/t/cost-of-vision-using-gpt-4o/775002 "2024-05-27T14:04:34Z")
I’m trying to calculate the cost per image processed using Vision with GPT-4o. I’m passing a series of jpg files as content in low detail: history = \[\] num\_prompt\_tokens = 0 num\_completion\_tokens = 0 num\_total\_tokens = … --- ## [Gpt4o ocr mistakenly bolds math symbol uvw (mathbf)](https://community.openai.com/t/gpt4o-ocr-mistakenly-bolds-math-symbol-uvw-mathbf/775076)
**Author:** [@ficapy](https://community.openai.com/u/ficapy)\ **Replies:** 5\ **Last updated:** [May 24, 2024, 3:31am UTC](https://community.openai.com/t/gpt4o-ocr-mistakenly-bolds-math-symbol-uvw-mathbf/775076 "2024-05-24T03:31:27Z")
I tried using gpt4o’s Vision API to recognize some mathematical literature and get LaTeX Markdown. In most cases, its accuracy reached a basically usable level. However, there is one particularly noticeable error: for u,… --- ## [Vision Support in Assistant API](https://community.openai.com/t/vision-support-in-assistant-api/590499)
**Author:** [@quentin.g](https://community.openai.com/u/quentin.g)\ **Replies:** 0\ **Last updated:** [January 17, 2024, 10:32am UTC](https://community.openai.com/t/vision-support-in-assistant-api/590499 "2024-01-17T10:32:13Z")
Is there any timeline guidance on when we might expect to see support for GPT-4-vision in the Assistant API? Also, when is vision due for general release (instead of preview)? We need to know for product planning and r… --- ## [Seeking Advice: Enhancing Accuracy of GPT-4 with Vision](https://community.openai.com/t/seeking-advice-enhancing-accuracy-of-gpt-4-with-vision/507814)
**Author:** [@HenriqueMelo](https://community.openai.com/u/HenriqueMelo)\ **Replies:** 5\ **Last updated:** [May 15, 2024, 7:32am UTC](https://community.openai.com/t/seeking-advice-enhancing-accuracy-of-gpt-4-with-vision/507814 "2024-05-15T07:32:24Z")
Hello everyone, I’ve been using the OpenAI API for some basic testing until now, and I’m planning to integrate it into a project of mine. Here’s the situation: I have a large set of multiple-choice questions, each accom… --- ## [Getting server-error on gpt-4 vision api](https://community.openai.com/t/getting-server-error-on-gpt-4-vision-api/733926)
**Author:** [@justgamingthings436](https://community.openai.com/u/justgamingthings436)\ **Replies:** 2\ **Last updated:** [May 2, 2024, 4:55pm UTC](https://community.openai.com/t/getting-server-error-on-gpt-4-vision-api/733926 "2024-05-02T16:55:26Z")
Hi, getting server\_error since last 12 hours in gpt-4 vision , { error: { message: 'The server had an error processing your request. Sorry about that! You can retry your request, or contact us through an Azure support… --- ## [I managed to build such an interesting scene from three photos + an additional object](https://community.openai.com/t/i-managed-to-build-such-an-interesting-scene-from-three-photos-an-additional-object/719509)
**Author:** [@ikereinez](https://community.openai.com/u/ikereinez)\ **Replies:** 0\ **Last updated:** [April 16, 2024, 4:34pm UTC](https://community.openai.com/t/i-managed-to-build-such-an-interesting-scene-from-three-photos-an-additional-object/719509 "2024-04-16T16:34:10Z")
This is a really impressive level for the DALL-E 3, considering that the entire scene was built from just three photos + one object Prompt Create a wide image where a whimsical scene unfolds in a magical forest. In the… --- ## [Am I using the latest version of GPT Vision?](https://community.openai.com/t/am-i-using-the-latest-version-of-gpt-vision/715680)
**Author:** [@maxm1](https://community.openai.com/u/maxm1)\ **Replies:** 2\ **Last updated:** [April 12, 2024, 9:50am UTC](https://community.openai.com/t/am-i-using-the-latest-version-of-gpt-vision/715680 "2024-04-12T09:50:32Z")
Hi, I have seen reference to ‘GPT4 Turbo Vision’ and struggling to work out what the latest version of the GPT4 Vision API to use. Right now, I am calling ‘gpt-4-vision-preview’ from my code, and the header response ret… --- ## [Azure OpenAI Platform Vs API](https://community.openai.com/t/azure-openai-platform-vs-api/703181)
**Author:** [@pathikghugare](https://community.openai.com/u/pathikghugare)\ **Replies:** 0\ **Last updated:** [March 31, 2024, 3:02pm UTC](https://community.openai.com/t/azure-openai-platform-vs-api/703181 "2024-03-31T15:02:42Z")
I’m using a GPT4, vision-preview deployment in my Azure OpenAI Platform in WEST-US-2 region I am having difficulties in identifying issues w.r.t Azure OpenAI API and Azure OpenAI Platform Consider following scenarios, … --- ## [Is there a limit of how many images you can upload to 'gpt-4-vision-preview'?](https://community.openai.com/t/is-there-a-limit-of-how-many-images-you-can-upload-to-gpt-4-vision-preview/698320)
**Author:** [@maxm1](https://community.openai.com/u/maxm1)\ **Replies:** 1\ **Last updated:** [March 26, 2024, 8:40pm UTC](https://community.openai.com/t/is-there-a-limit-of-how-many-images-you-can-upload-to-gpt-4-vision-preview/698320 "2024-03-26T20:40:33Z")
I’m running into an issue uploading a series of images to the GPT Vision API. I have compressed my images, and also added the ‘details:low’ parameter to the payload to make sure I stay within rate limits. Here is my sce… --- ## [Is the gpt-4-vision-preview no longer available?](https://community.openai.com/t/is-the-gpt-4-vision-preview-no-longer-available/547071)
**Author:** [@james9446](https://community.openai.com/u/james9446)\ **Replies:** 7\ **Last updated:** [March 6, 2024, 11:37am UTC](https://community.openai.com/t/is-the-gpt-4-vision-preview-no-longer-available/547071 "2024-03-06T11:37:25Z")
I’ve been working with the gpt-4-vision-preview for a couple of weeks now and it’s been working great. However, today all of my requests started failing. I’ve tested requests with the exact same structure and even retrie… --- ## ['gpt-4-vision-preview' always starting it's response with the same word](https://community.openai.com/t/gpt-4-vision-preview-always-starting-its-response-with-the-same-word/641259)
**Author:** [@maxm1](https://community.openai.com/u/maxm1)\ **Replies:** 1\ **Last updated:** [February 20, 2024, 10:39am UTC](https://community.openai.com/t/gpt-4-vision-preview-always-starting-its-response-with-the-same-word/641259 "2024-02-20T10:39:34Z")
Hi, I’ve tried changing various different ‘temperature’ values but no matter what I do the vision api is always starting the description of my images with the same word - “Nestled”. The rest of the response is different… --- ## [\[FEB 14\] GPT-4V and OpenAI Outage](https://community.openai.com/t/feb-14-gpt-4v-and-openai-outage/625617)
**Author:** [@Innovatix](https://community.openai.com/u/Innovatix)\ **Replies:** 1\ **Last updated:** [February 14, 2024, 7:24am UTC](https://community.openai.com/t/feb-14-gpt-4v-and-openai-outage/625617 "2024-02-14T07:24:52Z")
Many individuals are mistakenly equating GPT-4V with GPT-5, but there is no direct connection between the two. GPT-4V is actually referred to as gpt-4-vision-preview. https://help.openai.com/en/articles/8555496-gpt-4… --- ## [Is gpt-4-1106-vision-preview also getting lazy](https://community.openai.com/t/is-gpt-4-1106-vision-preview-also-getting-lazy/611557)
**Author:** [@utsav15.goel](https://community.openai.com/u/utsav15.goel)\ **Replies:** 6\ **Last updated:** [February 5, 2024, 7:44pm UTC](https://community.openai.com/t/is-gpt-4-1106-vision-preview-also-getting-lazy/611557 "2024-02-05T19:44:27Z")
Hi Community, I have a quick question. Since Openai has released a new GPT-4-Turbo model( gpt-4-0125-preview ), which solves the laziness problem, I was wondering if gpt-4-vision was also facing the same problem. I was… --- ## [Question: GPT Vision + Appending to the messages object for sequential requests](https://community.openai.com/t/question-gpt-vision-appending-to-the-messages-object-for-sequential-requests/601710)
**Author:** [@whispersofagi](https://community.openai.com/u/whispersofagi)\ **Replies:** 6\ **Last updated:** [February 2, 2024, 6:56pm UTC](https://community.openai.com/t/question-gpt-vision-appending-to-the-messages-object-for-sequential-requests/601710 "2024-02-02T18:56:57Z")
I am using the gpt-4-vision-preview model to analyse an image and I have some questions about forming sequential requests. I may need to reprompt the model with a different question depending on what the model’s response… --- ## [GPT-Vision - item location, JSON response, performance](https://community.openai.com/t/gpt-vision-item-location-json-response-performance/599194)
**Author:** [@georg-san](https://community.openai.com/u/georg-san)\ **Replies:** 0\ **Last updated:** [January 24, 2024, 12:48am UTC](https://community.openai.com/t/gpt-vision-item-location-json-response-performance/599194 "2024-01-24T00:48:45Z")
Request for features/improvements: Performance improvements (response time for 50-200 tokens on a 1MB image takes 10-15 seconds with detail parameter set to low Item location - API to include location data of items in … [Next page](https://community.openai.com/tag/gpt4-vision/323.md?match_all_tags=true&page=1&tags%5B%5D=gpt4-vision)