Medical Information Response Audit (MIRA): leaderboard
Large language models (LLMs) are increasingly used to provide public-facing health information, yet existing safety evaluations overlook whether responses preserve comparable medical information across different user phrasings of the same question. To address this, we introduce the Medical Information Response Audit (MIRA), a bilingual, controlled benchmark that assesses whether LLMs provide compa
Metric: Underinformative simplification (D3) (self-reported). Source: benchmarklist.com. Saturation forecast: Around January 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Pro | 2.16 | #102 |
| 2 | Qwen 3.6 Plus | 2.25 | #120 |
| 3 | GPT-5.4 | 2.29 | #79 |
| 4 | Claude Sonnet 4.6 | 2.32 | #90 |
| 5 | Llama 3.3 70B | 2.95 | #579 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medical-information-response-audit-mira · How It Works · Data refreshed daily, snapshot 2026-10-20.