← Back to all agents
Research Freemium

Elicit

Research agent that reads academic papers and extracts findings.

Our take

Indispensable for literature reviews — turns 50 papers into a table.

What it is

Elicit is a research assistant that searches across 200M+ academic papers and uses language models to extract key findings into structured tables. It is designed for empirical research workflows: literature reviews, systematic reviews, and evidence synthesis. You can ask a research question and get back a table of papers with their methods, sample sizes, and conclusions filled in.

Best for

  • Literature reviews and systematic reviews
  • Extract data from many papers into a comparison table
  • Find papers by methodology or outcome (not just keywords)

Pros

  • Built specifically for academic research workflows
  • Saves hours of manual paper skimming
  • Transparent about which papers the answers come from

Cons

  • Less useful outside empirical research (CS, biomed, social science)
  • Free tier is limited in number of papers per month

Quick start

  1. Sign up at elicit.com with your email or Google account — free tier gives a limited number of paper analyses per month.
  2. Type a research question (e.g., "Does meditation reduce anxiety in adults?"). Elicit returns a columnar table of real papers with extracted answers, sample sizes, and methodologies.
  3. Filter columns by sample size, intervention type, or outcome to drill into the evidence base.
  4. Click any row to read the source excerpt Elicit used, with the citation pinned to the sentence.
  5. Plus $12/mo and Pro $49/mo add more analyses per month, larger extracts, and alerts on new papers matching your query.

Sample input / output

Input
Research question: "Does mindfulness-based therapy reduce anxiety symptoms in college students? Show me RCTs from the last 5 years."
Output
Returned 18 RCTs as a structured table:
  - Hofmann et al. (2018, meta-analysis, n=2,341) — Hedge's g = 0.42, p<.001
  - Galante et al. (2021, Lancet Psychiatry, n=6,126) — effect size 0.32
  - Zhang et al. (2024, BMC Public Health, n=1,847 students) — g = 0.51
  ...
Inline excerpts link back to PDF pages. Click "Add to Library" to keep the search for later. 14/18 papers found on PubMed / Semantic Scholar verified.

Benchmarks

Indexed papers 200M+ across PubMed, Semantic Scholar, OpenAlex
Best fit Literature reviews, systematic reviews, evidence synthesis
Plus tier $12/mo, ~10× the free analyses
Pro tier $49/mo, full-text PDFs, alerts, larger extracts

Pricing

Free tier; Plus $12/mo

Underlying models

Custom fine-tuned models + GPT-4o class LLMs

Should you pick this?

Pick it if You're writing a literature review or systematic review and need to turn 50+ papers into a structured comparison table — Elicit was literally built for this workflow.
Skip it if Your question is outside empirical research (engineering, CS architecture questions, creative writing) — Consensus or Perplexity serve better there.

Similar agents