IBA-Bench - Medical Services: leaderboard

Metric: Checkpoint satisfaction (%; share of hidden preference checkpoints satisfied by the final action, medical services domain; mean of three LLM judges: DeepSeek-V3.2, Qwen3-4B-Instruct and Qwen3-30B-A3B-Instruct; retrieval-augmented baseline with bge-m3 top-3 chunks). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.

Top models

#ModelScore
1Qwen 3 4B Instruct82.9
2GPT-5 Mini80.2
3QwQ-32B79.6
4DeepSeek V3.279
5DeepSeek V378.8
6GPT-5.178
7Qwen 2.5 7B Instruct75
8GPT-4o73
9GPT-4o Mini71.8
10GLM-4 9B Chat69.3

Interactive version: theaggregate.ai/benchmark?slug=iba-bench-medical-services · How It Works · Data refreshed daily, snapshot 2026-09-26.