Record Test rt-v1.1 - Audio - Turn Boundary F1: leaderboard

Record Test rt-v1.1 - Audio - Turn Boundary F1. Exploratory: two fresh repetitions per model (reps 2-3 headline); failed draws stay in the denominator; no confidence intervals. Metric: Turn Boundary F1 (%). Higher is better.

Metric: Turn Boundary F1 (%). Source: heydonna.law. Saturation forecast: Estimated already saturated. 3 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 4 Argon (Low)91.62#11 (Gemini 4 Argon)
2Gemini 3.8 Flash (Low)91.06#36 (Gemini 3.8 Flash)
3Gemini 3.1 Flash Lite (Minimal)88.03#364 (Gemini 3.1 Flash Lite)

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=record-test-rt-v1-1-audio-turn-boundary-f1 · How It Works · Data refreshed daily, snapshot 2026-11-13.