How to Send Screenshots to ChatGPT Effectively
Get better answers from ChatGPT when sharing screenshots of errors, dashboards and documents: cropping, annotation and text extraction tips.
Crop to the question
A full-desktop screenshot forces the model to find the relevant part. Crop to the error dialog, chart, or table you are asking about, and the answer gets more specific.
Pair image with text
For error messages, logs and code, run the image through Screenshot to Text and paste the text too. Models cannot misread text you typed in, and you can search it later.
Say what you want
'What is wrong here?' is weaker than 'This error appears when I run npm build on Node 20. What causes it and how do I fix it?' Context turns a guess into a diagnosis.
Try it free in your browser
- Screenshot to Text — Extract readable text from any image with browser-side OCR.
Tools mentioned
Frequently asked
Should I send the image or the text?
Both when accuracy matters. The image gives layout, the text guarantees exact wording for error messages and code.
Can ChatGPT read tiny text in screenshots?
Small or compressed text is often misread. Zoom in before capturing or extract the text with OCR.
Keep reading
How to Prepare PDFs for ChatGPT, Claude, and Gemini
A practical guide to extracting clean, AI-ready text from PDFs — born-digital and scanned — so ChatGPT, Claude, and Gemini answer accurately and don't waste tokens on headers, footers, and page numbers.
The Best Way to Feed Long Documents to Claude (and Other Long-Context Models)
Claude's 200K-token context is generous, but you'll still want to clean, compress, and structure long documents before sending them. Here's a step-by-step playbook.
Cleaning Whisper Transcripts for AI Summaries
OpenAI Whisper, Otter, and YouTube transcripts are full of timestamps, filler words, and speaker noise. Here's how to strip them before sending to ChatGPT or Claude — and why it matters.