In this video, I will use to compare the Deep Research modes from different AI providers: ChatGPT, Grok, Perplexity, and DeepSeek. I will demonstrate how the results generated by ChatGPT are going much more in-depth while the results generated by Perplexity and DeepSeek are more dispersed, which can be more interesting if you just want to get a wide overview of a topic. Grok seems to be the most focused on the central concepts but once you remove the top layer of ideas it’s very dispersed, so it’s probably more similar to standard modes in other models.
You can run this same analysis using the InfraNodus extension: — it works for all major AI chatbots.
Timecodes:
0:00 Introduction: Deep Research
0:25 How we’re going to analyze Deep Research modes using knowledge graphs
1:43 How does Deep Research mode work and what is it?
3:58 ChatGPT Deep Research: very detailed and relatively focused
5:58 Grok Deep Research: too biased on the central concepts
8:15 Perplexity: highly diverse responses, focused on use cases and tools
9:00 DeepSeek: too dispersed and not so detailed, but good overview
10:49 Overview so far: what are the different modes good for?
11:36 💡 Special trick #:1 remove the top layer of ideas to zoom into underlying topics
16:32 💡 Special trick #2: generating prompts using the knowledge graph
18:13 💎 How InfraNodus makes discourse more coherent by connecting it more
19:20 For power users: switching from entity processing to word processing
source
Disclaimer
The content published on this page is sourced from external platforms, including YouTube. We do not own or claim any rights to the videos embedded here. All videos remain the property of their respective creators and are shared for informational and educational purposes only.
If you are the copyright owner of any video and wish to have it removed, please contact us, and we will take the necessary action promptly.
