Medi

Permanent benchmark record

Medi Bench

A versioned framework for comparing grounded workflow quality, accepted-workflow cost, and response speed as models change. Rules are fixed within each benchmark version.

Displayed evidence
V1 archived
Latest candidate
V3 pilot
Integrity
Ed25519 verified

PILOT / NOT FOR PUBLIC CLAIMS. The July 2026 V3 methodology is recorded, but no V3 model metrics are published. Its 260 cases, 3 repeats, and 2,340 observations are development/pilot process counts only, not comparative performance claims. The verified May 2026 V1 snapshot remains the last-known-good public evidence.

Last-known-good evidence

Medi Bench V1

The archived chart did not publish corpus size, repeats, confidence intervals, secondary rates, or prompt-level material.

Medi end-to-end benchmark comparisonGrounded answer rate on the vertical axis and Cost per accepted answer (USD) on the horizontal axis. Larger bubbles indicate faster p95 latency. Use arrow keys to move between models.Cost per accepted answer (USD)Grounded answer rate

V1 reported grounded-answer rate, cost per accepted answer, and p95 latency. It is archived directional evidence with no published corpus size, repeats, or confidence intervals, and is not directly comparable with the planned V3 weighted methodology.

Equivalent table for the end-to-end chart
ModelGrounded answer rateCost per accepted answer (USD)p95 latencyConfidenceRun
99%$0.00409.7sThe archived aggregate fixture did not publish confidence intervals.Not published · Not published prompts · No repeats published
100%$0.00088.1sThe archived aggregate fixture did not publish confidence intervals.Not published · Not published prompts · No repeats published
98%$0.003510.6sThe archived aggregate fixture did not publish confidence intervals.Not published · Not published prompts · No repeats published
95%$0.000514.3sThe archived aggregate fixture did not publish confidence intervals.Not published · Not published prompts · No repeats published

Methodology

Fixed rules before results.

The contract fixes track definitions, missing-value behaviour, public-data boundaries, and publication stops before any candidate can appear on the chart.

Scoring

  • V1 reported grounded-answer rate, cost per accepted answer, and p95 latency.
  • The archive uses only the exact rounded values visible in the May 2026 chart; missing information remains null.

Public-data boundary

  • Public artifacts are aggregate-only. Frozen prompts, source content, model responses, provider identifiers, and prompt-level scores remain private.
  • Publication fails closed until public-private separation and human review are complete.

Model eligibility

Planned is not published.

V3 records the approved comparison matrix. Public eligibility remains pending until access, pricing, completeness, and review are represented in a signed aggregate artifact.

July 2026 V3 model matrix
ModelTierPublic status
GPT-5.4 miniTier 1 candidatePending / not published
GPT-5.6 LunaTier 1 candidatePending / not published
GPT-5 miniTier 1 candidatePending / not published
GPT-4.1 miniTier 1 candidatePending / not published
GPT-5.6 TerraTier 2 comparatorPending / not published
GPT-5.6 SolTier 2 comparatorPending / not published
GPT-5.4Tier 2 comparatorPending / not published
GPT-5.5Tier 2 comparatorPending / not published

Tracks

One model. Two questions.

01

Oracle-context

Measures model behaviour with the approved evidence context supplied directly.

  • The oracle-context track isolates model behaviour by supplying the approved evidence context directly.
  • It reports no result until the frozen corpus, grading, and human review gates are complete.

02

End-to-end

Measures the complete Medi retrieval, policy, grounding, citation, and answer path, including failures and latency.

  • The end-to-end track includes retrieval, policy, model response, citation selection, failures, cost, and latency.
  • Failed observations remain in aggregate outcome and latency accounting and cannot be deleted to improve publication status.

Confidence intervals

Uncertainty belongs beside the score.

  • V3 specifies prompt-clustered bootstrap confidence intervals so repeats do not count as independent prompts.
  • No interval is published until the corresponding reviewed aggregate metric is published.

The archived V1 fixture did not publish intervals or repeat counts. The interface says so directly rather than implying precision that the source aggregate does not support.

Limitations

  • PILOT / NOT FOR PUBLIC CLAIMS: V3 has no reviewed public model result artifact in this snapshot.
  • The approved model matrix is a benchmark plan, not evidence of provider access, model availability, or performance.
  • Model availability, pricing, adapters, retrieval, and infrastructure conditions may change after capture.
  • Benchmark evidence does not replace human review or authorize autonomous medical, legal, employment, or final compliance decisions.

Historical versions

History only moves forward.

  1. 2026-07-09

    Medi Bench V3

    PILOT / NOT FOR PUBLIC CLAIMS. Methodology recorded; model results not published.

  2. 2026-05-27

    Medi Bench V1

    Verified archived aggregate and current last-known-good public evidence.

Change log

What changed.

V3

Recorded the two-track V3 methodology, publication stop rules, and approved model matrix without publishing model metrics.

V1

Retained the archived aggregate V1 snapshot as the last known good public evidence.

Machine-readable snapshots

Each response includes the parsed snapshot and its Ed25519 integrity metadata. The index is append-only and hash-chained.

Trust by design

Built for regulated care teams

Compliance Health is designed for high-trust environments: Canadian residency, accountable automation, and clear evidence trails.

Visit Trust Centre

Canada-hosted

Customer data stored and processed in Canada (AWS ca-central-1).

Human oversight

Human review for consequential, ambiguous, low-confidence, exception, and escalation outcomes.

Audit-ready evidence

Evidence trails and logs designed for inspections and internal governance.

ISO/IEC 42001 roadmap

AI management controls in place; certification roadmap in progress.