Auto-posted whereas I am in Tokyo. Operating these checks 24/7 on VPS.
I have been working the identical Gold buying and selling prompts by three totally different AI fashions for every week. Identical account, similar professional advisor (DoIt Alpha Pulse AI), fully totally different considering patterns.
This is what’s truly occurring with Claude, GPT-5, and Gemini after they analyze Gold.
The Take a look at Setup (You Can Replicate This)
The Precise Immediate I am Utilizing
Present XAUUSD: [price] Final 3 H1 candles: [data] Session: [London/NY/Asian] Information at this time: [economic calendar] Ought to I: Purchase/Promote/Maintain? Threat: 0.5% max Goal: Threat-reward 1:2 minimal Clarify reasoning in 50 phrases max.
Easy. Clear. Identical for all three fashions.
Testing Circumstances
- Demo account: $5000
- Every mannequin will get: $1500 allocation
- Identical trades provided: All three see equivalent setups
- Determination tracked: Even after they say “Maintain”
- Time recorded: Response pace issues
Early Observations (Not Conclusions)
GPT-5: The Overthinker
Response time: 3-5 seconds
GPT-5 retains discovering patterns that may not exist. Yesterday it mentioned:
“The three-candle formation resembles the Could 2023 reversal sample mixed with present DXY weak point suggesting institutional accumulation nonetheless the amount profile signifies…”
Drawback: By the point it finishes considering, the entry is gone.
Fascinating conduct: It catches delicate correlations. Seen that Gold was ignoring Greenback energy as a result of bond yields had been additionally rising. That is truly refined.
Present standing:
- Alerts generated: 12
- Trades taken: 4 (others too gradual)
- Win fee: 50% (2 wins, 2 losses)
- P&L: +45 pips
Claude Opus 4.1: The Pace Dealer
Response time: 1-2 seconds
Claude makes choices FAST. Typically too quick. Its responses are like:
“Bullish. London open + assist held + Greenback weak. Purchase.”
Power: In quick markets, Claude truly will get fills. Throughout Wednesday’s volatility, it was the one mannequin that caught the reversal.
Weak point: Much less nuanced. Missed the Bond/Gold correlation fully.
Present standing:
- Alerts generated: 18
- Trades taken: 11
- Win fee: 54% (6 wins, 5 losses)
- P&L: +72 pips
Gemini 2.5: The Conservative One
Response time: 2-4 seconds (varies)
Gemini is extra cautious. Typically passes on trades the others take. Tuesday it mentioned:
“No clear edge. Recommend ready for higher setup.”
This occurs extra with Gemini than GPT or Claude.
Surprising energy: Threat administration. When unsure, it usually suggests smaller positions. The one mannequin that often says “cut back threat to 0.25%” when confidence is decrease.
Minor weak point: Typically TOO conservative, lacking good strikes whereas ready for “good” setups.
Present standing:
- Alerts generated: 9
- Trades taken: 5
- Win fee: 60% (3 wins, 2 losses)
- P&L: +38 pips
The Fascinating Discovery: They Typically Disagree
More often than not, they agree on course. However this is what occurred Thursday at London open:
Gold worth: 1952.30
Setup: Break above Asian excessive
- GPT-5: “Watch for pullback to 1950”
- Claude: “Purchase now, momentum constructing”
- Gemini: “Purchase however smaller place”
Identical bullish bias, totally different approaches to entry.
Claude entered instantly. Gold ran to 1958. Claude obtained the very best entry.
However all three would have been worthwhile – simply totally different quantities.
What’s Truly Priceless Right here
Pace vs Intelligence Commerce-off
- Want quick choices? Claude
- Want deep evaluation? GPT-5
- Want threat administration? Gemini (surprisingly)
Value Per Determination (This Week)
- GPT-5: $0.12 common
- Claude: $0.08 common
- Gemini: $0.06 common
Claude is 33% cheaper AND sooner. However GPT-5’s two wins had been greater (+40 and +35 pips vs Claude’s common of +20).
The “Confidence” Drawback
None of those fashions say “I do not know” sufficient. They all the time have an opinion, even after they should not.
I am testing including this to prompts:
If unclear, say "No edge - skip this setup" Confidence required: 70% minimal
Early outcomes: 40% fewer alerts, however higher win fee.
The Framework That is Rising
After one week, this is what I am studying:
Use Claude When:
- Information is about to hit (pace issues)
- London/NY session opens (momentum trades)
- You want fast choices on clear setups
Use GPT-5 When:
- Asian session (extra time to assume)
- Complicated correlations matter
- You may watch for good entries
Use Gemini When:
- You need a second opinion
- Threat administration is precedence
- Testing new methods (it is extra conservative)
What’s Truly Working Nicely
Easy Operations
One factor that stunned me – DoIt Alpha Pulse AI handles all three fashions with out points:
- No API errors (correct error dealing with in-built)
- No fee restrict issues (clever request administration)
- Constant connections throughout all fashions
That is truly our aggressive benefit. Whereas others battle with integration, we simply… commerce.
The Actual Variations Are Delicate
The fashions are extra comparable than totally different. All of them:
- Catch fundamental assist/resistance
- Perceive development course
- React to main information
The variations are in model, not substance:
- Claude: Direct and quick
- GPT-5: Detailed and considerate
- Gemini: Cautious and measured
The “Clarification Tax”
Asking for reasoning provides:
- 1-2 seconds to response time
- 2x the token price
- Typically overthinking easy setups
However it’s price it for studying what the AI “sees”
What I am Testing Subsequent Week
Experiment 1: Consensus Buying and selling
Solely take trades the place 2 of three fashions agree. Concept: Greater conviction setups.
Experiment 2: Time-Based mostly Rotation
- Asian: Gemini (conservative for quiet markets)
- London: Claude (pace for breakouts)
- NY: GPT-5 (complexity of US session)
Experiment 3: Specialised Prompts
As a substitute of 1 immediate for all, optimize for every mannequin’s strengths:
- Claude: Quick, action-focused
- GPT-5: Embody correlation evaluation
- Gemini: Add threat parameters
The Sincere Actuality
After one week of parallel testing, the fashions carry out equally on Gold buying and selling.
All of them catch the apparent strikes. The variations are marginal – possibly 5-10% efficiency variance. The talent is not selecting the “proper” AI – it is writing higher prompts.
That is why DoIt Alpha Pulse AI helps all of them. Not as a gimmick, however as a result of totally different market circumstances want several types of considering.
Your Homework Whereas I am in Japan
In case you have DoIt Alpha Pulse AI, do that:
- Run the identical setup by totally different fashions
- Doc after they disagree
- Observe which one was proper
- Share findings
By the point I am again, we’ll have crowd-sourced information on which mannequin works greatest for what.
The Questions I am Investigating in Tokyo
Assembly with quant merchants right here who’ve been utilizing AI longer:
- How do they deal with mannequin disagreement?
- What’s their method to consensus?
- How do they optimize for latency from Asia?
- Are there fashions we’re not contemplating?
Present Scoreboard (Week 1)
Pace Champion: Claude (1-2 seconds)
Accuracy Chief: Gemini (60% win fee however small pattern)
Complexity Grasp: GPT-5 (catches delicate patterns)
Value Winner: Gemini ($0.06/choice)
Reliability: Claude (most constant)
However bear in mind – that is one week of knowledge. Not conclusions, simply observations.
The Actual Worth of This Experiment
It isn’t about discovering the “greatest” mannequin. It is about understanding that AI buying and selling technique is not one-size-fits-all.
Your buying and selling model, the pairs you commerce, your threat tolerance – all of them have an effect on which AI mannequin fits you.
That is why the immediate is extra essential than the mannequin. An excellent immediate on Claude beats a foul immediate on GPT-5 each time.
Wish to run your personal AI mannequin experiments?
Get DoIt Alpha Pulse AI – Now $397
Helps all main AI fashions. Change between them immediately. Discover what works for YOUR buying and selling.
P.S. – Nonetheless in Tokyo. These fashions are working 24/7 on my VPS. Once I test in from my lodge, I see Claude and GPT-5 arguing about whether or not 1958 is resistance or assist. Even AIs cannot agree on fundamental TA.
P.P.S. – If you happen to’re testing fashions your self, doc all the things. The patterns solely emerge with information, not hunches.
