By skyforbes I switch between ChatGPT, Claude, and Perplexity depending on the task, but for research-heavy questions I keep gravitating toward Perplexity.
The difference for me isn’t just that it shows citations; it’s how the answers are structured around them.
The difference for me isn’t just that it shows citations; it’s how the answers are structured around them.
When I asked a medical question in all three tools, the Perplexity answer came back in sections, with each part tied to specific sources listed alongside or underneath.
That layout made it easier to see which claims came from which references, almost like a lightweight research summary rather than a single blended response.
ChatGPT and Claude can also provide sources, especially in their browsing/research modes, but in my experience their default formatting feels less tightly coupled to specific citations unless I prompt very explicitly.
I’m curious how much of this comes from Perplexity’s retrieval and answer pipeline versus its UI and formatting choices and why ChatGPT doesn't perform the same…