BacktestBench - Ticker Selection: leaderboard

The Ticker Selection task family of BacktestBench (2026): selecting the stock a natural-language strategy would pick, 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 Ticker Selection task family (selecting the stock a natural-language strategy would pick), 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

#ModelScoreOverall rank
1Gemini 3 Pro90.37#145
2Seed 1.885.83#252
3Qwen 3 Max85.56#415
4Qwen 3 235B A22B 2507 (Thinking)82.26#466
5MiniMax-M2.181.71#575
6GLM-4.780.33#380
7DeepSeek V3.279.78#419
8Qwen 3 32B75.93#972
9GPT-OSS-120B74#704
10MiMo-V2-Flash73.04#684
11Qwen 3 30B A3B 2507 (Thinking)71.25#764
12Qwen 3 Next 80B A3B (Thinking)67.4#594
13Kimi K2 (Thinking)66.57
14Qwen 3 Coder Plus63.41#720
15Qwen 3 14B62.04#1269

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=backtestbench-ticker-selection · How It Works · Data refreshed daily, snapshot 2026-11-21.