Skip to content

ScienceAi2 (Allen Institute for AI)

Asta

Open agentic research assistant for scientists, from Ai2.

Categories
ScienceResearch
Pricing
FREE
Hosting
Cloud
Platforms
WebAPI
Models
Multi-model
Verified
Jun 20, 2026

Asta is an open ecosystem of AI agents for science from the Allen Institute for AI (Ai2). The Asta assistant helps researchers find relevant papers, generate cited literature summaries, and (in beta) run basic data analyses, with planned hypothesis generation and experiment replication. It ships alongside AstaBench, a benchmark suite for evaluating scientific agents.

Pros & cons

  • Cited literature summaries with sources
  • Open ecosystem plus the AstaBench benchmark
  • Built by a respected nonprofit AI lab (Ai2)
  • Rankings weigh cost, not just accuracy
  • Data analysis is in limited beta
  • Narrower than general-purpose research assistants
  • Focused on scholarly/scientific workflows

Tags

Further reading

View all Science
  • View FutureHouse details
    ScienceFREE

    FutureHouse

    FutureHouse

    AI agents that automate scientific literature research and discovery.

    FutureHouse is a nonprofit AI-for-science lab building AI agents that automate steps of the research process. Its free FutureHouse Platform exposes specialized agents — including Crow, Falcon, Owl, and Phoenix — for literature search, synthesis, and chemistry, plus Kosmos, an 'AI Scientist' that runs autonomous multi-step analyses. The agents are available through a web app and an open API.

    Free to use
    Narrow to research workflows
    • research
    • science-agents
    • literature
    • nonprofit
  • View Elicit details
    ResearchFREEMIUM

    Elicit

    Elicit (formerly Ought)

    AI research assistant for literature review and evidence synthesis.

    Searches, summarizes, and extracts structured data across 125M+ academic papers, with sentence-level citations. Goes beyond chat-over-papers with a guided systematic-review workflow covering search, screening, extraction, and report synthesis. A free Basic plan offers unlimited search and summaries; paid tiers add data-extraction volume and the review pipeline.

    Systematic-review screening pipeline
    No PICO framework or Boolean operators
    • research
    • literature-review
    • systematic-review
    • citations
  • View Consensus details
    ResearchFREEMIUM

    Consensus

    Consensus

    AI search over 200M+ peer-reviewed papers, with citations.

    An academic search engine that runs natural-language questions against a corpus of more than 200 million peer-reviewed papers and synthesizes the findings with inline citations. Its Consensus Meter aggregates how much the studies agree or disagree, and a copilot helps draft literature reviews. Aimed at researchers, students, and clinicians.

    200M+ peer-reviewed corpus
    Best answers behind paid tier
    • research
    • academic-search
    • citations
    • literature-review
  • View Undermind details
    ResearchFREEMIUM

    Undermind

    Undermind

    AI deep-research agent that finds every relevant paper on a scientific question.

    A research-discovery platform that takes a natural-language question and autonomously searches, reads, and evaluates hundreds of academic papers — following citation trails and explaining each result's relevance, mimicking how an expert runs a literature review. Built by MIT quantum-physics PhDs and backed by Y Combinator; used by researchers at MIT, Harvard, and Caltech.

    Exhaustive citation-trail search
    Slow — minutes per query
    • research
    • literature-search
    • science
    • deep-research