GPT-4o vs Qwen3-32B in finance

Every finance benchmark where both have a score (OpenAI and Qwen). Data refreshed 2 Nov 2026.

Shared benchmarks
2
GPT-4o only
4
Qwen3-32B only
1

Shared benchmarks, side by side

Each benchmark is drawn on its own scale. Bars are not comparable across rows.

  • GPT-4o
  • Qwen3-32B

  • BizFinBench average score across 9 tasks, higher is better

    GPT-4o:71.8 (better)
    Qwen3-32B:68.3

  • K-FinHallu binary detection F1, higher is better

    GPT-4o:0.733 (better)
    Qwen3-32B:0.732
Benchmark
GPT-4o
Qwen3-32B

8 benchmarks, 2 scored for both. Bold marks the better score where both have one.

BizFinBenchaverage score across 9 tasks, higher is better
71.8
Paper Table 3 (Reported by the authors, 26 May 2025)
68.3
Paper Table 3 (Reported by the authors, 26 May 2025)
K-FinHallubinary detection F1, higher is better
0.733
Paper Table 3 (Reported by the authors, 28 May 2026)
0.732
Paper Table 3 (Reported by the authors, 28 May 2026)
FinanceQAaccuracy (total, 0-1), higher is better
0.392
Paper Table 1 (Reported by the authors, 30 Jan 2025)
Not scored
FAITHoverall accuracy %, higher is better
Not scored
73.9
Paper Table 3 (Reported by the authors, 24 Oct 2025)
FinLFQAevidence attribution F1, higher is better
56.5
Paper Table 6 (Reported by the authors, 7 Oct 2025)
Not scored
FinLFQALLM-as-judge answer score (/15), higher is better
13.5
Paper Table 6 (Reported by the authors, 7 Oct 2025)
Not scored
InvestorBenchcumulative return %, higher is better
39.0
Paper Table 2 (Reported by the authors, 24 Dec 2024)
Not scored
Open FinLLM Leaderboardbalanced score (mean of text/vision/audio averages), higher is better
46.0
Open FinLLM Leaderboard (scores.json) (Official board, 16 Apr 2026)
Not scored

Scores from different benchmarks use different scales and are not added up. Settings can differ between sources; see each leaderboard for details.