0

Trading Bot Bench

Frontier

Single-turn python algorithmic trading script creation.

Domain
rl-env
License
unknown
Published
Sep 2025

Cite

Notes

Only stored in your browser.

Attribution

Leaderboard scores
prime-hub
Attribution policy →

Top score 39.8% by GPT-4.1 Nano - 4 models reporting (4 frontier)

Score history

4
0%25%50%75%100%Apr 25May 25Jun 25Jul 25Aug 25GPT-4.1 MiniGPT-4.1 Nano

Top models

4
Trading Bot BenchBar chart with 4 bars. Highest value: GPT-4.1 Nano at 39.8.
4 models

Related tools

1
View all

Implementations, trainers, datasets and scaffolds linked to this eval.

FAQ

What is Trading Bot Bench?
Single-turn python algorithmic trading script creation.
What is the current top score on Trading Bot Bench?
The top reported score is 39.8% by GPT-4.1 Nano, across 4 models reporting (4 from frontier labs).
How can a model improve its Trading Bot Bench score?
Tools linked to Trading Bot Bench on Sophon include BOT Bench RL Env (Community) - RL environments, datasets, and scaffolds that target this eval.
What license is Trading Bot Bench under?
Trading Bot Bench is available under unknown.