Nicholai/jan
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Added links for individual deep research pieces

Browse Source
This commit is contained in:
Daniel Ching 2025-08-08 20:57:24 +08:00
parent b423d7291e
commit 700af88bf7
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
docs/src/pages/post/deepresearch.mdx
View File

@ -77,27 +77,27 @@ Generate a comprehensive report about the state of AI in the past week. Include
new model releases and notable architectural improvements from a variety of sources. new model releases and notable architectural improvements from a variety of sources.
``` ```
Google's generated report was the most verbose, with a whopping 23 pages that reads [Google's generated report](https://github.com/menloresearch/prompt-experiments/blob/main/Gemini%202.5%20Flash%20Report.pdf) was the most verbose, with a whopping 23 pages that reads
like a professional intelligence briefing. It opens with an executive summary, like a professional intelligence briefing. It opens with an executive summary,
systematically categorizes developments, and provides forward-looking strategic systematically categorizes developments, and provides forward-looking strategic
insights—connecting OpenAI's open-weight release to broader democratization trends insights—connecting OpenAI's open-weight release to broader democratization trends
and linking infrastructure investments to competitive positioning. and linking infrastructure investments to competitive positioning.
OpenAI produced the most citation-heavy output with 134 references throughout 10 pages [OpenAI](https://github.com/menloresearch/prompt-experiments/blob/main/OpenAI%20Deep%20Research.pdf) produced the most citation-heavy output with 134 references throughout 10 pages
(albeit most of them being from the same source). (albeit most of them being from the same source).
Perplexity delivered the most actionable 6-page report that maximizes information [Perplexity](https://github.com/menloresearch/prompt-experiments/blob/main/Perplexity%20Deep%20Research.pdf) delivered the most actionable 6-page report that maximizes information
density while maintaining scannability. Despite being the shortest, it captures all density while maintaining scannability. Despite being the shortest, it captures all
major developments with sufficient context for decision-making. major developments with sufficient context for decision-making.
Claude produced a comprehensive analysis that interestingly ignored the time constraint, [Claude](https://github.com/menloresearch/prompt-experiments/blob/main/Claude%20Deep%20Research.pdf) produced a comprehensive analysis that interestingly ignored the time constraint,
covering an 8-month period from January-August 2025 instead of the requested week (Jul 31-Aug covering an 8-month period from January-August 2025 instead of the requested week (Jul 31-Aug
7th 2025). Rather than cataloging recent events, Claude traced the evolution of trends over months. 7th 2025). Rather than cataloging recent events, Claude traced the evolution of trends over months.
Grok produced a well-structured but relatively shallow 5-page academic-style report that [Grok](https://github.com/menloresearch/prompt-experiments/blob/main/Grok%203%20Deep%20Research.pdf) produced a well-structured but relatively shallow 5-page academic-style report that
read more like an event catalog than strategic analysis. read more like an event catalog than strategic analysis.
Kimi produced a comprehensive 13-page report with systematic organization covering industry developments, research breakthroughs, and policy changes, but notably lacks proper citations throughout most of the content despite claiming to use 50-100 sources. [Kimi](https://github.com/menloresearch/prompt-experiments/blob/main/Kimi%20AI%20Deep%20Research.pdf) produced a comprehensive 13-page report with systematic organization covering industry developments, research breakthroughs, and policy changes, but notably lacks proper citations throughout most of the content despite claiming to use 50-100 sources.
### Understanding Search Strategies ### Understanding Search Strategies