diff --git a/features/agent.mdx b/features/agent.mdx
index dcae11ec..948a8975 100644
--- a/features/agent.mdx
+++ b/features/agent.mdx
@@ -186,6 +186,10 @@ Pass the `model` parameter to select which model to use:
 - **Data Gathering**: "Extract contact information from company websites"
 - **Content Summarization**: "Summarize the latest blog posts about web scraping"
 
+## Large Result Sets
+
+A single agent run can typically return around **100–200 rows** of structured tabular data. This is due to the output token capacity of the underlying language model, not a plan restriction. If your task requires more data, split it into multiple focused agent runs — for example, one run per geographic region or category — and merge the results on your end. Keeping your schema consistent across runs makes combining the output straightforward.
+
 ## CSV Upload in Agent Playground
 
 The [Agent Playground](https://www.firecrawl.dev/app/agent) supports CSV upload for batch processing. Your CSV can contain one or more columns of input data. For example, a single column of company names, or multiple columns such as company name, product, and website URL. Each row represents one item for the agent to process.