TruelineAutoevalSign in
    ResearchWebinarsBlogAutoeval DocsHelp & SupportLearn EvalsComing Soon
    Pricing
    About UsContact

    PLUMLOOM / AUTOEVAL / FREE PILOT

    Free Autoeval Pilot

    Limited to ten teams in the pilot

    We're offering complimentary hands-on onboarding, evaluation support, and model access for up to 10 teams. Applications are considered in the order received, so sign up early.

    Open Source · Apache-2.0

    Multiple-Trial Evidence

    Score / 5
    Average Score
    4.21 / 5
    95% Confidence Interval
    3.82–4.60
    Release Bar: 4.0
    3.55.0
    INCONCLUSIVEIllustrative

    Development Workflow

    Claude CodeCodexCursor...
    CLIMCPCI

    Model Providers

    OpenAIAnthropic (Coming Soon)Together.ai
    MicrosoftVisaIBMSiemens HealthineersAlteryxClarivateAllstateOptumTyler TechnologiesSeismicInfo EdgeMahindra Last Mile MobilityathenahealthAvalaraRealPageKPMG

    01 / THE PILOT

    From Setup to Release Gates

    2 Hours of Hands-On SupportModel Access IncludedNo Purchase Commitment
    01

    Evaluation Setup

    Help configuring Autoeval and preparing an evaluation for your team's use case.

    02

    Run Evaluations

    Assistance with single-run or multiple-trial evaluations using supported models, including Plumloom-provided model access.

    03

    Review Results

    Scores, judge agreement, variation, confidence intervals where applicable, and possible failures.

    04

    Release Gating

    Help configuring a PASS / FAIL / INCONCLUSIVE release gate for your evaluation.

    02 / SCOPE

    Focused Support

    Up to 2 hours of complimentary support per team, typically one onboarding session and one evaluation/results-review session. Delivered remotely within 30 days of pilot confirmation.

    Model usage is subject to an agreed pilot allowance. No unlimited credits.

    Setup, guidance, execution assistance, and results review only. Custom application development, major integrations, and ongoing support are outside the free pilot.

    No purchase or paid subscription commitment.

    03 / YOUR APPLICATION

    Sign Up for the Free Pilot

    Please don't submit API keys, passwords, proprietary source code, or sensitive evaluation artifacts.

    About You
    Your Evaluation Needs
    What Are You Building or Evaluating?
    Which Evaluation Types Interest You?
    Which Modalities Are Relevant to Your Application?

    Current Autoeval evaluations cover text scenarios, conversation transcripts, and agent traces. Voice / Audio and Images / Vision are future capabilities, not currently available. Your selections help us understand interest.

    Services and Training

    Interest only. No purchase commitment.

    04 / FAQ

    Quick Answers

    Who is this for?

    Teams building AI agents, RAG assistants, and LLM-powered applications. Free Autoeval Pilot: Limited to 10 Teams. Applications are considered in the order received, so sign up early.

    Do we need to provide model keys?

    No. Plumloom-provided model access is included within an agreed pilot allowance. Never submit provider API keys in this form.

    Does it cost anything?

    No. The pilot includes up to 2 hours of complimentary hands-on support. No purchase or paid subscription commitment is required.

    What happens after we apply?

    We'll contact you by email with next steps.