Research agents Hacker News commenters recommend
5 research agents were recommended on Hacker News between November 2025 and October 2026. We rank them by how many separate people recommend each one.
None of the 5 research agents earns the badge. The most recommended is Autoresearch by Karpathy: 25 recommend it and 13 criticize it.
All 5 research agents, ranked
Select a column heading to sort. Mixed counts the people with mixed feelings, who count toward neither side. Approval share is recommenders out of recommenders and critics; the tick on each bar marks the 2 to 1 line the badge needs. Months counts the months, out of 12, in which someone recommended it. Had for is the median time people say they have had or used it, shown when 10 or more said.
| Rank | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | Autoresearch by Karpathy | 25 | 9 | 13 | 66% | 6 | — | — |
| 2 | Gemini Deep Research | 20 | 3 | 17 | 54% | 8 | — | — |
| Tied 3 | OpenAI Deep Research | 7 | 5 | 13 | 35% | 6 | — | — |
| Tied 3 | Tongyi DeepResearch | 7 | 0 | 0 | 100% | 1 | — | — |
| 5 | Kagi Research | 6 | 2 | 0 | 100% | 6 | — | — |
How we rank
Rank is the number of separate people who recommend a product in a Hacker News comment between November 2025 and October 2026. Someone who praises the same research agent in five threads counts once, in the column of their latest opinion.
A research agent earns the badge when at least 10 people recommend it, recommenders outnumber critics at least 2 to 1 (mixed feelings count toward neither side), it was recommended in at least 4 of the 12 months and in at least 5 different threads, and no single thread supplies more than half of the people who recommend it.
Rankings are never for sale. We may earn a commission when you buy through our links. It never changes the order.
Counts come from Hacker News comments labeled by Claude, with data through October 31, 2026. Counts updated 2 Nov 2026. How it works