CaReCoS - ZCH: leaderboard

Metric: Accuracy (%) of one-to-two-sentence answers to clinically grounded questions about a mel-spectrogram image of the recording, judged by an LLM against the reference (strict factual matching for explicit questions, relaxed semantic matching for inferred ones); the 80 ZCH neonatal and paediatric heart-sound questions; higher is better. Source: arxiv.org. Saturation forecast: Around February 2028. 9 models tracked.

Top models

#ModelScore
1GPT-5.131.2
2Gemini 2.5 Pro28.7
3Gemini 2.5 Flash27.5
4Qwen2.5-Omni-7B15

Interactive version: theaggregate.ai/benchmark?slug=carecos-zch · How It Works · Data refreshed daily, snapshot 2026-09-29.