BankerToolBench Verifier - F1: leaderboard
BankerToolBench Verifier - F1. Source TableA10,judging same expert-labeled taskset and ground truth. Judge models are evaluated as verifiers,not task-solving agents. Cost/latency excluded. Metric: F1 (0–1). Primary published table; source configurations remain distinct.
Metric: F1 (0–1). Source: arxiv.org. Saturation forecast: Estimated already saturated. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 0.91 | #89 |
| 2 | Gemini 3 Flash (Preview) | 0.9 | #80 |
| 3 | Gemini 3 Pro (Preview) | 0.88 | #64 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=bankertoolbench-verifier-f1 · How It Works · Data refreshed daily, snapshot 2026-10-16.