LakeQA - Mini: leaderboard
Metric: End-to-end exact match (%) on the 135 LakeQA-mini tasks, a stratified subset of LakeQA-full that preserves the reasoning-intensity distribution; the agent answers a multi-hop question over a data lake of about 40 million Data.gov and Wikipedia documents with one tool call per round (search, list, download, inspect, query) until it answers or hits the turn limit; exact match against the annotator-verified answer; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.5 | 45.93 |
| 2 | Claude Sonnet 4.5 | 33.33 |
| 3 | GPT-5.2 | 17.04 |
| 4 | Claude Haiku 4.5 | 11.11 |
| 5 | DeepSeek R1 | 6.67 |
| 6 | GPT-5 Mini | 2.22 |
| 7 | Llama 3.3 70B Instruct | 1.48 |
Interactive version: theaggregate.ai/benchmark?slug=lakeqa-mini · How It Works · Data refreshed daily, snapshot 2026-09-29.