I am testing some OCR image-to-text parsing with GPT-4o using both the ChatGPT UI and the OpenAI chat.completions.create endpoint. I have a few questions I would like to get your help and input on. I am trying to understand why the ChatGPT UI is performing much better at extracting information fr…

I struggled with this and eventually found the API sample code I lifted from docs had max-tokens set to 300 (at the bottom of the json). I changed to 4095 and received the full response.

Parse image to text with gpt-4o with ChatGpt UI and OpenAI chat.completions.create endpoint - Very Different Results

API

nicholishen July 30, 2024, 3:01pm 3

Do you guys have any sample images?

Topic		Replies	Views
Make OpenAI Vision API Match GPT4 Vision API chatgpt	4	3821	December 6, 2023
The performance difference between ChatGPT4o and gpt4o api using the same prompt for image analysis API gpt-4 , chatgpt , gpt-4-vision , gpt4-vision , api-vision	5	1033	July 27, 2024
GPT-4 omni text recognition via API works worse than on chatgpt.com API gpt-4 , api	4	1192	August 13, 2024
Incomplete Output of 4o through API API gpt-4o	3	856	May 20, 2024
Can an assistant help me with OCR? API gpt-4	7	3432	June 6, 2024

Parse image to text with gpt-4o with ChatGpt UI and OpenAI chat.completions.create endpoint - Very Different Results

Related topics