BacktestBench - Metrics Calculation: leaderboard
The Metrics Calculation task family of BacktestBench (2026): computing a backtest metric such as return or drawdown from a natural-language strategy, from a strategy written in natural language over a database of real market records. Each model drives the same retrieval, SQL and Python-execution pipeline; the score is accuracy in percent against the executed reference backtest; higher is better.
Metric: Back-test accuracy (%) on the Metrics Calculation task family (computing a backtest metric such as return or drawdown from a natural-language strategy), synthetic BacktestBench test split (the expert-crafted subset excluded), each model driving the AutoBacktest pipeline of the authors (Summarizer, SQL Retriever and Python Coder agents) over a PostgreSQL market database, temperature 0.6; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 23 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 51.67 | #145 |
| 2 | Qwen 3 Max | 44.14 | #415 |
| 3 | Seed 1.8 | 43.22 | #252 |
| 4 | GLM-4.7 | 39.13 | #380 |
| 5 | Qwen 3 235B A22B 2507 (Thinking) | 34.71 | #466 |
| 6 | GPT-OSS-120B | 30.86 | #704 |
| 7 | Kimi K2 (Thinking) | 30.02 | |
| 8 | DeepSeek V3.2 | 29.18 | #419 |
| 9 | Qwen 3 30B A3B 2507 (Thinking) | 28.67 | #764 |
| 10 | Qwen 3 32B | 27.93 | #972 |
| 11 | Qwen 3 Next 80B A3B (Thinking) | 26.02 | #594 |
| 12 | MiniMax-M2.1 | 21.98 | #575 |
| 13 | Qwen 3 Coder Plus | 20.59 | #720 |
| 14 | MiMo-V2-Flash | 17.43 | #684 |
| 15 | GPT-OSS-20B | 17.33 | #1183 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=backtestbench-metrics-calculation · How It Works · Data refreshed daily, snapshot 2026-11-21.