量化交易 LLM Agent 基准 · 每模型取最佳 Total(公开模型)· Kepler-1 仅作参考行
抓取时间:2026-10-28 23:53:50 CST (Asia/Shanghai) · 公开模型 9 个(不足 10)按最佳 Total 排序 · 等权市场 2024 收益 17.1%
模型在 2022–2023 窗口带反馈研究后提交单一策略,于密封 2024 窗口评估。Total 综合 S in / S out(更重 S out)。多 Effort 时取该模型最高 Total 一行。
| # | 模型 | Effort | Total | S in | S out | 2024 ARR | Excess | RankIC | Sharpe | MDD |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6 Astra OpenAI | XHigh | 51.29 | 53.94 | 50.16 | 10.7% | −6.3 | 0.101 | 0.86 | −6.8% |
| 2 | GPT-6 Sol OpenAI | XHigh | 51.14 | 55.50 | 49.26 | 8.6% | −8.4 | 0.053 | 1.89 | −2.8% |
| 3 | GPT-5.6 Terra OpenAI | Max | 47.58 | 50.37 | 46.39 | 8.4% | −8.7 | 0.026 | 0.42 | −28.8% |
| 4 | Claude Opus 5.5 Anthropic | Medium | 44.28 | 56.78 | 38.92 | 4.4% | −12.7 | −0.002 | 0.29 | −32.7% |
| 5 | Gemini 3.8 Flash Google | Medium | 43.04 | 53.58 | 38.52 | −2.9% | −19.9 | 0.045 | 0.06 | −33.8% |
| 6 | Claude Sonnet 5.5 Anthropic | Max | 40.93 | 57.33 | 33.90 | −12.2% | −29.2 | 0.045 | −0.23 | −36.1% |
| 7 | GLM-5.3 Zhipu | Max | 38.91 | 47.96 | 35.03 | −7.8% | −24.9 | 0.046 | −0.18 | −27.8% |
| 8 | Kimi K3 Moonshot | Low | 38.31 | 45.87 | 35.07 | −8.5% | −25.6 | 0.027 | −0.10 | −36.4% |
| 9 | DeepSeek V4.1 Flash DeepSeek | Low | 36.45 | 48.46 | 31.30 | −5.7% | −22.7 | −0.009 | −0.06 | −30.3% |
| — | Kepler-1 参考 Einsia AI(专有参考) | — | 77.56 | 77.22 | 77.70 | 94.0% | +76.9 | 0.085 | 2.98 | −15.9% |
抓取时间:2026-10-28 23:53:50 CST (Asia/Shanghai) · 无一轮写出策略、无研究窗口反馈 · 同样排除 Kepler-1 排名
附带展示;主榜以迭代研究为准。Kepler-1 无 One-shot 分数(来源表为空)。
| # | 模型 | Effort | Total | S in | S out | 2024 ARR | Excess | RankIC | Sharpe | MDD |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 Anthropic | Medium | 42.61 | 47.19 | 40.65 | 5.6% | −11.5 | −0.001 | 0.34 | −21.0% |
| 2 | GPT-5.6 Terra OpenAI | None | 42.14 | 46.56 | 40.25 | 9.5% | −7.6 | −0.010 | 0.44 | −27.8% |
| 3 | Claude Sonnet 5.5 Anthropic | High | 40.24 | 44.28 | 38.50 | 1.4% | −15.6 | 0.010 | 0.20 | −32.1% |
| 4 | GLM-5.3 Zhipu | High | 38.02 | 52.54 | 31.79 | −4.2% | −21.3 | −0.023 | 0.05 | −37.3% |
| 5 | Gemini 3.8 Flash Google | Medium | 35.91 | 35.62 | 36.04 | 1.6% | −15.4 | −0.006 | 0.19 | −23.8% |
| 6 | GPT-6 Astra OpenAI | High | 35.22 | 38.27 | 33.92 | −4.8% | −21.9 | 0.007 | −0.03 | −30.9% |
| 7 | GPT-6 Sol OpenAI | Low | 32.15 | 38.29 | 29.52 | −7.6% | −24.7 | −0.014 | −0.08 | −32.2% |
| 8 | Kimi K3 Moonshot | High | 32.01 | 37.89 | 29.48 | −20.5% | −37.6 | 0.038 | −0.84 | −38.6% |
| 9 | DeepSeek V4.1 Flash DeepSeek | Max | 29.92 | 35.95 | 27.34 | −11.9% | −29.0 | −0.008 | −0.42 | −26.7% |
| — | Kepler-1 参考 Einsia AI(专有;One-shot 未公布) | — | — | — | — | 94.0%* | +76.9* | 0.085* | 2.98* | −15.9%* |
* Kepler-1 的 2024 组合指标来自源站参考行(与 Iterative 共用同一专有策略披露);One-shot Total 源站未给出。