FinInteract: leaderboard

Metric: Accuracy (%; all 173 bilingual instances; ReAct agent in the standard full-interaction mode (A+S+I): it may search the filing corpus, ask the simulated user (GPT-5, answering yes, no or I don't know from the hidden disambiguating context) yes-or-no questions, and answer; a final answer counts as correct only if it matches the intended interpretation, graded by GPT-4o-mini with a 1% numeric tolerance). Source: arxiv.org. Saturation forecast: Around 2037. 9 models tracked.

Top models

#ModelScore
1Qwen 3.5 35B A3B28.9
2Qwen 3 8B24.9
3Qwen 3 14B22.5
4GPT-520.2
5Qwen 3 4B12.1
6Qwen 3 30B A3B11.6
7Qwen 3 32B8.4
8GPT-4o4.6
9GPT-5 Mini0

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