summarize-document · gpt-4o · writing · 90% confidence
Summarize a document with GPT-4o — optimized prompt
A GPT-4o summarization prompt: fixed audience and length, extractive not interpretive, delimiters around the source, no preamble.
90% confidenceUpdated 9/7/2026summarize-document · gpt-4o · writing
Original prompt
Summarize this document for me: [PASTE DOCUMENT]
Optimized prompt
Summarize the document below for a specific reader. Be extractive — report what the text says, do not add interpretation or outside context. Reader: [role, e.g. "a busy exec deciding whether to fund this"] Max length: [e.g. 150 words / 5 bullets] <document> [PASTE DOCUMENT] </document> Output: - The summary, at or under the length limit. - "Not stated:" one line flagging anything the reader will want that the document does not actually cover. No preamble, no "this document discusses", no restating the instructions.
Why this GPT-4o prompt works
Tuned for GPT-4o: a named reader, a hard length cap, and an extractive-only rule. GPT-4o is concise and obedient to a word limit, so the prompt spends its instructions on what NOT to do — no interpretation, no ‘this document discusses’ opener.
What to check in the output
- The optimized_prompt should consistently produce summaries that are more focused on the core message and key supporting details than the vibe_prompt.
- Summaries generated by the optimized_prompt should be less prone to including tangential information or irrelevant details.
- The optimized_prompt should lead to summaries that are generally more consistent in quality and structure across different documents.