GPT-5-Thinking vs Numa 1 in finance
Every finance benchmark where both have a score (OpenAI and Moneda). Data refreshed 2 Nov 2026.
- Shared benchmarks
- 1
- GPT-5-Thinking only
- 0
- Numa 1 only
- 6
Shared benchmarks, side by side
Each benchmark is drawn on its own scale. Bars are not comparable across rows.
- GPT-5-Thinking
- Numa 1
FinSearchComp Global subset average accuracy %, higher is better
GPT-5-Thinking:63.9 (better)Numa 1:59.4FinSearchComp Greater China subset average accuracy %, higher is better
GPT-5-Thinking:46.4Numa 1:57.1 (better)
Benchmark
GPT-5-Thinking
Numa 1
8 benchmarks, 2 scored for both. Bold marks the better score where both have one.
FinSearchCompGlobal subset average accuracy %, higher is better
FinSearchCompGreater China subset average accuracy %, higher is better
FinanceQAaccuracy (total, 0-1), higher is better
Not scored
FinanceBench% correct, no filing given (150-question sample), higher is better
Not scored
Moneda CiteBenchFactual share of linked claims %, higher is better
Not scored
Moneda FilingsBench% correct (private split), higher is better
Not scored
Moneda PortfolioBenchPortfolio tasks, % correct, higher is better
Not scored
Finance Agent BenchmarkPartial credit % on the 27 public questions (run by Moneda; not comparable, unranked), higher is better
Not scored
Scores from different benchmarks use different scales and are not added up. Settings can differ between sources; see each leaderboard for details.