FinED-Bench - Financial Domain Knowledge Errors: leaderboard

Metric: F1 (%; financial domain knowledge errors: terminology misuse, incorrect legal reference, ambiguous expression, numerical unit error, omitted financial element; a predicted error counts as correct when the extracted sentence matches or contains the annotated sentence and the error type is classified correctly; the paper's error-detection prompt over each document, reasoning models with thinking on unless marked no thinking). Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked.

Top models

#ModelScore
1GPT-4o47.36
2Qwen 3 14B41.1
3Qwen 3 8B38.6
4Qwen 3 14B (Non-reasoning)26.63
5DeepSeek R1 0528 Qwen3 8B22.82
6Qwen 3 8B (Non-reasoning)18.06
7Qwen 2.5 7B Instruct8.33
8GPT-4o Mini7.34
9GPT-3.5 Turbo3.65

Interactive version: theaggregate.ai/benchmark?slug=fined-bench-financial-domain-knowledge-errors · How It Works · Data refreshed daily, snapshot 2026-09-26.