FinSearchComp: leaderboard

635 financial search questions from 70 finance professionals on global and Greater China markets: time-sensitive data fetching, historical lookup and historical investigation; ByteDance Seed, 2025.

Metric: Average Score (%). Source: randomtutu.github.io. Status: saturation imminent. 12 models tracked.

Top models

#ModelScore
1Grok 468.9
2GPT-5 (Thinking)63.9
3Gemini 2.5 Pro42.6
4Qwen 3 235B A22B 250737.4
5Kimi K229.5
6DeepSeek R117.2

Interactive version: theaggregate.ai/benchmark?slug=finsearchcomp · How It Works · Data refreshed daily, snapshot 2026-09-05.