DelistBench (Web Search): leaderboard
Metric: Overall joint accuracy within 7 days (%; all 1,200 security-level delisting records: a Category A record needs the correct positive status, the announcement date within seven days and the broad reason, Category B and C records a correct negative status; invalid outputs and abstentions count as errors; web-enabled, provider search). Source: arxiv.org. Saturation forecast: Around December 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.6 Flash | 81.5 |
| 2 | GPT-5.6 Luna (Low) | 79.4 |
| 3 | DeepSeek V4 Pro (High) | 79 |
| 4 | DeepSeek V4 Flash (Non-reasoning) | 78.3 |
| 5 | Gemini 3.5 Flash Lite (Medium) | 75.9 |
Interactive version: theaggregate.ai/benchmark?slug=delistbench-web-search · How It Works · Data refreshed daily, snapshot 2026-09-26.