"Can AI trade crypto? Jay Azhang, a computer engineer and finance bro from New York, is putting this question to the test with Alpha Arena. The project pits the greatest large language models (LLM) against each other, each with 10 thousand dollars worth of capital, to see which can make more money trading crypto. The models include Grok 4, Claude Sonnet 4.5, Gemini 2.5 pro, ChatGPT 5, Deepseek v3.1, and Qwen3 Max."
"three out of the five AIs are underwater, with Qwen3 and Deepseek - the two Chinese open source models - leading the charge. That's right, the western world's most powerful, closed source, proprietary artificial intelligences run by giants like Google and OpenAI, have lost over $8,000 dollars, 80% of their crypto trading capital in little over a week, while their eastern open source counterparts are in the green."
Alpha Arena runs a head-to-head crypto trading contest where large language models each receive $10,000 to trade. Competitors include Grok 4, Claude Sonnet 4.5, Gemini 2.5 Pro, ChatGPT 5, Deepseek v3.1, and Qwen3 Max. Early results show three of five AIs losing money, while Qwen3 and Deepseek, two Chinese open-source models, lead with gains. Western proprietary models managed significant losses—over $8,000, roughly 80% of the trading capital—within about a week. Notable trades include Qwen3's 20x Bitcoin long and Grok 4's prolonged 10x Dogecoin long. Gemini maintained broadly bearish, short positions.
Read at Bitcoin Magazine
Unable to calculate read time
Collection
[
|
...
]