QuantArena - One-shot - Research Period: leaderboard
QuantArena One-shot: shared equity research environment,2022-2023 research and sealed 2024 out-of-sample windows. Version-fixed quality/diversity composite scaled0-100; out-of-sample combination unpublished. Each of 38 requested effort configurations retained; no best-effort pooling. These evaluate agent-generated strategies, not investment advice or model response speed. Score (%).
Metric: Score (%). Source: arxiv.org. Saturation forecast: Not forecast. 38 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5.3 (High) | 52.54 | #36 (GLM-5.3) |
| 2 | Claude Opus 5.5 (xHigh) | 49.05 | #4 (Claude Opus 5.5) |
| 3 | Claude Sonnet 5.5 (xHigh) | 47.9 | #11 (Claude Sonnet 5.5) |
| 4 | Claude Opus 5.5 (Medium) | 47.19 | #4 (Claude Opus 5.5) |
| 5 | GPT-5.6 Terra (Non-reasoning) | 46.56 | #40 (GPT-5.6 Terra) |
| 6 | Claude Sonnet 5.5 (High) | 44.28 | #11 (Claude Sonnet 5.5) |
| 7 | Claude Opus 5.5 (Low) | 43.21 | #4 (Claude Opus 5.5) |
| 8 | GPT-5.6 Terra (xHigh) | 42.94 | #40 (GPT-5.6 Terra) |
| 9 | GPT-5.6 Terra (High) | 42.88 | #40 (GPT-5.6 Terra) |
| 10 | Claude Sonnet 5.5 (Low) | 42.41 | #11 (Claude Sonnet 5.5) |
| 11 | Claude Opus 5.5 (Max) | 42.07 | #4 (Claude Opus 5.5) |
| 12 | Claude Sonnet 5.5 (Medium) | 41.81 | #11 (Claude Sonnet 5.5) |
| 13 | Claude Sonnet 5.5 (Max) | 41.19 | #11 (Claude Sonnet 5.5) |
| 14 | Kimi K3 (Max) | 40.87 | #30 (Kimi K3) |
| 15 | GPT-6 Astra (Max) | 40.13 | #6 (GPT-6 Astra) |
No result here: #7 GPT-6.1 Sol, #8 Gemini 4 Argon, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=quantarena-one-shot-research-period · How It Works · Data refreshed daily, snapshot 2026-10-23.