TORNEO
Menu

We run the AI tools you are considering on the same real task, under a protocol locked in advance, and publish what happened.

Each tool's estimate with its uncertainty, the cost and the date. When the data cannot separate the tools, we say so instead of ranking them.

Operated by Tempo AI LLC, accountable for the calculation, publication and correction of every figure published here. No vendor pays to be measured; TORNEO buys what it tests.

done · 2 published categories. Each result keeps its date, uncertainty and limits.

Request a tournament · Follow the beta runs

Beta runs

partial · Played measurements; publication checks remain separate.

Latest case study · transcription ·

Faithfully transcribe audio into text (speech-to-text via API)

Measured with intervals across 2 replicated blocks; the intervals overlap, no tool is ranked first.

Interval bands, transcription, observed 2026-09-02T18:42:22.745775Z: gemini-3-5-transcribe interval from 0.420246 to 0.622379, estimate 0.522222; gemini-2-5-flash interval from 0.463641 to 0.664234, estimate 0.566667; gemini-2-5-pro interval from 0.507825 to 0.705301, estimate 0.611111; gemini-3-5-flash interval from 0.541502 to 0.735561, estimate 0.644444.overlapgemini-3-5-transcribe0.42 to 0.62gemini-2-5-flash0.46 to 0.66gemini-2-5-pro0.51 to 0.71gemini-3-5-flash0.54 to 0.740.30.40.50.60.70.8Interval bands, transcription, observed 2026-09-02T18:42:22.745775Z: gemini-3-5-transcribe interval from 0.420246 to 0.622379, estimate 0.522222; gemini-2-5-flash interval from 0.463641 to 0.664234, estimate 0.566667; gemini-2-5-pro interval from 0.507825 to 0.705301, estimate 0.611111; gemini-3-5-flash interval from 0.541502 to 0.735561, estimate 0.644444.overlapgemini-3-5-transcribe0.42 to 0.62gemini-2-5-flash0.46 to 0.66gemini-2-5-pro0.51 to 0.71gemini-3-5-flash0.54 to 0.740.30.40.50.60.70.8
Share of recordings transcribed with no error at all, estimate and interval per tool; the tinted zone is where every interval overlaps. Population n = 90, 2 replicated blocks, observed , valid until .

This is our experience, on our corpus, at this date. It is not buying advice.

Do not name a winner: the confidence intervals overlap. Report the range, the n and the observation date.

Read the full result

All results

  • transcriptionMeasured, no tool first

    4 participants measured with intervals across 2 replicated blocks, intervals overlap, no tool ranked first (as of 2026-09-02)

  • emailfindNo rank demonstrated

    2 participants measured, no rank exists at current precision (as of 2026-09-02)

How we work

The protocol is locked before the first call: task, metric, population, budget, stopping rules. The lock is public, and every result links to it.

Every estimate carries an interval. When intervals overlap, no rank is published, and the page says so in one sentence.

Every published number can be recomputed from the published files with one command. Vendors get a public right of reply, published word for word.

The method in seven principles · Who operates and funds this