🌌 LIMITLESS HQ: LISTEN & FOLLOW HERE ⬇️
——
In this episode of Limitless, we discuss algorithmic trading, showcasing AI models that doubled $10,000 investments in just two weeks. Highlighting six models with varying performances, DeepSeek and Qwen achieved over 100% returns, while others like ChatGPT struggled.
We discuss AI trading behaviors, ethical considerations, and the impact of blockchain transparency through platforms like Hyperliquid. As the experiment wraps up, listeners can look forward to more insights on the evolving integration of AI in finance.
——
TIMESTAMPS
0:00 AI Models Can Make You Rich
1:01 Winners and Losers in Trading
1:57 Analyzing DeepSeek’s Success
4:33 Insights from AI Trading Logs
8:49 Trading Models: Skill or Luck?
10:54 The Value of Public Data
17:20 How to Trade Like an AI
21:53 The Risks of Algo Trading
22:29 The Future of AI in Trading
——
RESOURCES
Josh:
Ejaaz:
——
Not financial or tax advice. See our investment disclosures here:
sooop.. is it grok or deepseek that works the bezt
HI, I want to do the same, can you help me?
what has happened since then for the last 3 months?
Why is Deepseek so sucessfull????
WOW a whole 2 weeks of data? Congrats on cherry picking the data that looks the best. There needs to be much much more testing before anyone should be trading with any of those strategies. To be fair they may still work over an extended period of time but nothing here in this video shows that. Take the setup logic and backtest it yourself.
So what's the update on these?
This aged well🤣🤣
It'll get better as it learns
I wonder how kimi k2 would do
Not gonna lie, the fact that 2/3 ai models either didn’t make money or lost money isn’t exactly encouraging lol.
I would defiinitely like to speak with you guys because I have been making code with various AI sites and they all fail in real world strategies. Now I know why so many bots have to lie and show only good trades because they don't do so good everyday. Also you can run test with the bots and depending on how you run the test you can show good winning percentage and then when you sale them you tell people in disclaimer that there is no guarantee because past scenarios don't guarantee wins in the future!!! What I will say is that when you take your trading knowledge with the use of AI you could possibly build good indicators to go buy instead of paying huge sums of money instead!
DeepSeek (alongside all the other models) fails to produce risk-adjusted returns greater than an S&P 500 buy-and-hold. 0 edge in giving machine learning models trading data, when they're trained to predict words, as they'll simply predict what the average person they're trained on would have said after being given the data, which provides no advantage unless the data you're giving it has predictive power. In the case of having predictive power, an LLM is still a horrible way of doing decision-making compared to other, more reliable quantitative methods. Their performance resembles a random walk more than anything else in this project
Assuming you used enough leverage to have as large swings in your portfolio as DeepSeek did, you would have produced even higher returns by simply doing beta capture on the US economy on avg. As one of the fundamental sayings in data science goes: Garbage in, garbage out. Feeding things like RSI into an LLM will produce random results in the end. LLMs do have use cases in trading, but it's very niche, and using them for decision-making in trades is not a good idea
It's still quite an interesting project though, so props to the person running it to see how things go, and to you guys for having a nice podcast about it. It was nice to listen to, and pretty well-made
How to start ? Where can i get source code?
what is the name of the platform?
How to build these models for myself?
Was the DeepSeek model using RAG and an API ?
They don't know how to get out when the market is in a downtrend. They're like buying all the way down. Anyone can make money in a bull market.
The problem is obvious… All of,these models are paper trading pure and simple. Easy to trade with no real risk, so why ever paper trade conservatively? Simply go with your skill set, combined with intuition and go for it.
Not so easy because you need to hold during downturns, which involve emotions that obviously, Ai models don’t have to deal with. So the human trader will not run parallel with DeepSeek at the end of,the day…
Gambling with superior knowledge btw which does somewhat contradict my thesis. ☑️👌😎 JP
Please, how can i copy the Deepseek trade in Hyperliquid?
After 14 days it was abruptly stopped as all Models were loosing money, some clearly heading towards 0. Don't be fooled by the hype. i hope you will give the real story in an updated video.
According to Deepseek, there is no DeepSeek Trading bot
does anyone know what website they were looking at in the first half of the video?
Kept bumping into comments about Nixorus books, everyone saying they're dangerously honest and almost banned-level info. Eventually, I caved and checked it out. They're right—this stuff hits different. It's weirdly addictive, probably because it feels like knowledge you're not supposed to find.
A lot of talking but not actually going through first how it did and what was the set up. Once explained then critique it. Nothing gained in watching this.