Comment by sethops1

> Testing GPT-5, Claude, Gemini, Grok, and DeepSeek with $100K each over 8 months of backtested trading

So the results are meaningless - these LLMs have the advantage of foresight over historical data.