Exa
90/100
Public-surface review of Exa
Research agents reviewed by Hlido
Analysts, researchers, product strategists, and knowledge teams buy Research agents to turn scattered sources into decisions without losing citations or freshness. The failure mode is confident synthesis from stale, irrelevant, or invented sources, which can contaminate a memo, investment view, or product roadmap. Hlido checks source transparency, citation paths, follow-up behavior, recency signals, exportability, and pricing for real research workflows. The category score reflects Strategic Alpha, Execution Grit, Craft & Soul, and Value Signal.
Browse all 26 →How we evaluate Research agents
We test source transparency, citation paths, freshness signals, follow-up behavior, exportability, and pricing for actual research workflows. The evidence package highlights whether synthesis is traceable enough for a team to rely on it.
Top Research picks right now
Current top 3 by Laddoo Score across the Research corpus.
Browse the full Research corpus
Sortable, filterable list with tier and last-tested date.
| Agent | Score | Tier | Finding | Last tested |
|---|---|---|---|---|
| Exa | 90/100 | VITAL | Public-surface review of Exa | 2026-05-01 |
| Elicit | 90/100 | VITAL | Public-surface review of elicit-research. | 2026-05-01 |
| Exa Search | 90/100 | VITAL | Public-surface review of exa-search-api. | 2026-05-01 |
| Mendable | 78/100 | STEADY | Public-surface review of Mendable | 2026-05-01 |
| Scite | 78/100 | STEADY | Public-surface review of Scite | 2026-05-01 |
| Scholarcy | 78/100 | STEADY | Public-surface review of Scholarcy | 2026-05-01 |
| Kagi | 78/100 | STEADY | Public-surface review of Kagi | 2026-05-01 |
| Hebbia | 78/100 | STEADY | Public-surface review of Hebbia | 2026-05-01 |
| Contextual AI | 78/100 | STEADY | Public-surface review of Contextual AI | 2026-05-01 |
| Elastic Quote | 78/100 | STEADY | Public-surface review of Elastic Quote | 2026-05-01 |
| Utopai-Research/pai-pro | 67/100 | FADING | Utopai-Research/pai-pro — Local AI filmmaking studio — skills, canvas, timeline — driven from your coding agent.. | 2026-07-16 |
| Glean | 65/100 | FADING | Public-surface review of Glean | 2026-05-01 |
| Notion AI | 65/100 | FADING | Public-surface review of Notion AI | 2026-05-01 |
| Elicit | 65/100 | FADING | Public-surface review of Elicit | 2026-05-01 |
| NotebookLM | 65/100 | FADING | Public-surface review of NotebookLM | 2026-05-01 |
| Pensieve | 60/100 | FADING | Pensieve should get a cautious T1 public-surface pass focused on visible claims, access friction, pricing clarity, | 2026-05-08 |
| Perplexity | 53/100 | FADING | AI-powered answer system with cited sources. Public no-login basic queries. Pro tier for advanced reasoning models. | 2026-05-01 |
| Consensus | 53/100 | FADING | Public-surface review of Consensus | 2026-05-01 |
| Andi Search | 53/100 | FADING | Public-surface review of Andi Search | 2026-05-01 |
| Komo | 53/100 | FADING | Public-surface review of Komo | 2026-05-01 |
| Jetwriter AI: Personalized AI Writing Assistant for Chrome, Desktop & Mobile. Free to Try. | 50/100 | FADING | Public-surface review of Jetwriter AI: Personalized AI Writing Assistant for Chrome, Desktop & Mobile. Free to | 2026-05-14 |
| Xiangyue-Zhang/auto-deep-researcher-24x7 | 50/100 | FADING | Public-surface review of Xiangyue-Zhang/auto-deep-researcher-24x7. | 2026-06-18 |
| zjunlp/SciAtlas | 50/100 | FADING | Public-surface review of zjunlp/SciAtlas. | 2026-07-16 |
| genspark | 40/100 | FADING | Public-surface review of genspark. | 2026-05-01 |
| askiva | 40/100 | FADING | Public-surface review of askiva. | 2026-05-01 |
| You.com Search | 40/100 | FADING | Public-surface review of You.com Search | 2026-05-01 |