Recognize text, data and key points from an image or screenshot in 2 minutes using Qwen and DeepSeek
Free analysis of images, screenshots and documents without installing software
- 1.Open Qwen Chat at chat.qwen.ai. Click «Sign In» and register using your Google account or email. After logging in, you will land in the chat window.
If the site is in Chinese - switch the language to English via the gear icon ⚙️ in the top right corner, then select English.
Qwen Chat ↗ - 2.In the message input field, click the paperclip icon 📎 (or image icon) to the left of the input bar. Select the screenshot or photo of the document you want to analyze from your computer or phone. Wait for the file to upload and a preview to appear.
Supported formats: JPG, PNG, WEBP. File size - up to several megabytes. Do not upload documents with confidential personal data - this is a third-party service.
Qwen Chat ↗ - 3.Paste the following prompt into the chat bar, then click send. Qwen will analyze the image and produce a structured result.Qwen Chat ↗Prompt
Analyze this image carefully. Do the following: 1. Extract ALL text visible in the image, preserving layout and structure. 2. Identify the type of document or screenshot (e.g., receipt, chart, table, code, UI mockup, handwritten note). 3. Summarize the key information in 3-5 bullet points. 4. If the image contains a table, reproduce it in Markdown format. Format your answer as: ## Document Type ... ## Extracted Text ... ## Key Points ... ## Table (if any) ...
What this prompt doesThis prompt asks the AI to: extract all text from the image, identify the document type, highlight key points, and reproduce tables in Markdown format. No replacements needed - the prompt is universal. If you need a Russian translation, add at the end: «Translate all output to Russian.» - 4.If you need to deeply analyze the recognized text (e.g., find anomalies in a table, summarize a contract, or prepare a response based on a document) - copy the result from Qwen, open chat.deepseek.com, create a new chat, and paste the prompt below along with the text from step 3.
DeepSeek currently does not accept images directly in the free web chat - text only. So first extract text via Qwen (steps 1 - 3), then analyze the text via DeepSeek.
DeepSeek Chat ↗PromptHere is text extracted from a document. Please: 1. Find any errors, contradictions, or unusual values. 2. Answer this question about the content: [insert your question here]. 3. Provide a concise summary in 3 sentences. --- EXTRACTED TEXT --- [insert here the text obtained from Qwen in step 3] --- END ---
What this prompt doesThe prompt asks DeepSeek to: find errors and inconsistencies in the text, answer your specific question, and provide a brief summary. Replace [insert your question here] with what you want to know (e.g., «What is the contract duration?»). Replace [insert here the text...] with the result from Qwen. - 5.Done. Copy the result you need (extracted text, table, or analysis) into your document or notes.
To copy a table into Excel - select the Markdown table, paste it into any Markdown-to-Excel converter (e.g., markdowntoexcel.com), or paste the text into Google Sheets and use «Data -> Split text to columns».
Browser ↗
- freeQwen ChatFree web chat, sign up via Google or email, no payment required; there are message limits, but for a few images per day it is more than enough.
- freeDeepSeek ChatFree web chat with no card required; works with text, has daily request limits.
Why today
AI services change fast - interfaces and free limits may differ from what's described.