• Welcome back! Thank you for being a part of this Traders Community. Let's discuss and share :)
    Selamat datang kembali! Trimakasih telah menjadi bagian dari Komunitas Trader ini. Mari berdiskusi dan berbagi :)

New EA DeepGold LLM EA – when gold listens to artificial intelligence (well, almost)

NOWASwe said:
2026.07.13 08:15:00.699 DeepGold LLM EA v1.3 (XAUUSD,M15) HTTP error: 402 response: {"error":{"message":"Insufficient Balance","type":"unknown_error","param":null,"code":"invalid_request_error"}}
Click to expand...
you need to top up your balance on platform.deepseek.com
 
Hi,
Result with pepperstone razor real account for the last 2 days using this set

1784492791139.png

sometime the position start positive and close with a loss due to the trailing activation profit at 400 points. But if i use a less activation profit like 300 the profit will be minus than actual. so need to see the evolution between profit and loss in a more long period.
 

Attachments

Hello, colleagues!

Now our "golden" AI advisor has stopped being stubborn as a sheep and has learned to obey his master.

What was added:

• Two new fields in the settings: LLM_BuyThreshold and LLM_SellThreshold

(search in the "AI Settings (DeepSeek)" group).

Previously, the thresholds were tightly sewn in:

prediction >= 0.65 → BUY, prediction <= 0.35 → SELL.

Now you decide for yourself with what confidence the AI should open a deal.

If you want, tighten the nuts to 0.8 and 0.2 in order to trade only "reinforced concrete" signals.

If you want, loosen to 0.6 and 0.4 and catch more movements (but with popcorn, because it can be stormy).

Now, instead of "the robot thinks you're a sucker and ignores your wishes," we have a "henpecked robot" - turn the sensitivity knob and watch as it obediently runs to open positions.

Try it, experiment, and share the results. Version 1.4 is already in the topic header. Profit for everyone!
 

Attachments

NPNavigator said:
Hello, colleagues!

Now our "golden" AI advisor has stopped being stubborn as a sheep and has learned to obey his master.

What was added:

• Two new fields in the settings: LLM_BuyThreshold and LLM_SellThreshold

(search in the "AI Settings (DeepSeek)" group).

Previously, the thresholds were tightly sewn in:

prediction >= 0.65 → BUY, prediction <= 0.35 → SELL.

Now you decide for yourself with what confidence the AI should open a deal.

If you want, tighten the nuts to 0.8 and 0.2 in order to trade only "reinforced concrete" signals.

If you want, loosen to 0.6 and 0.4 and catch more movements (but with popcorn, because it can be stormy).

Now, instead of "the robot thinks you're a sucker and ignores your wishes," we have a "henpecked robot" - turn the sensitivity knob and watch as it obediently runs to open positions.

Try it, experiment, and share the results. Version 1.4 is already in the topic header. Profit for everyone!
Click to expand...
Hi,

do you thing the trading session is important in set or we can use 0 to 24 ?
 
Ludo69 said:
Hi,

do you thing the trading session is important in set or we can use 0 to 24 ?
Click to expand...
If you set it from 0 to 24, the adviser will turn into a coffee-drinking workaholic who, even in the Asian session, will look for signals where gold is sleeping like a bear in a den. Seriously— the session is very important for XAUUSD. The main movers (and normal spreads) are London and New York. The Asian session is usually "thin," with random spikes and wild spreads that AI can mistake for the turn of the century. As a result, false entries and draining of the deposit. Therefore, I would leave a reasonable range (for example, 9-23 UTC) — let the robot rock the gold while life is in full swing, and at night let it also sleep like a normal trader. But if your soul requires experimentation, no one forbids you to put 0-24 and see what your deposit will turn into in a week, then tell me! In general, a session is not just for show, but a real noise filter.
 
NPNavigator said:
If you set it from 0 to 24, the adviser will turn into a coffee-drinking workaholic who, even in the Asian session, will look for signals where gold is sleeping like a bear in a den. Seriously— the session is very important for XAUUSD. The main movers (and normal spreads) are London and New York. The Asian session is usually "thin," with random spikes and wild spreads that AI can mistake for the turn of the century. As a result, false entries and draining of the deposit. Therefore, I would leave a reasonable range (for example, 9-23 UTC) — let the robot rock the gold while life is in full swing, and at night let it also sleep like a normal trader. But if your soul requires experimentation, no one forbids you to put 0-24 and see what your deposit will turn into in a week, then tell me! In general, a session is not just for show, but a real noise filter.
Click to expand...
Thank for your answer. it's exactly what i thing due to my today experience. i change it to UTC 2 - 20 and i have 3 big loss .
1784738676400.png

1784738696130.png

So i'll change it to euope en us time like 8 to 20 .
what about the trailing activation. to secure the profit i change it to 250 .

the api send which informations to deepseek for analysis ?

regards
 
Ludo69 said:
Thank for your answer. it's exactly what i thing due to my today experience. i change it to UTC 2 - 20 and i have 3 big loss .
View attachment 172172

View attachment 172173

So i'll change it to euope en us time like 8 to 20 .
what about the trailing activation. to secure the profit i change it to 250 .

the api send which informations to deepseek for analysis ?

regards
Click to expand...
Three big losses in the Asian session is like going to a club at 7 a.m. hoping to meet a party. Gold is usually asleep at this time, and if it wakes up, it's only to draw some game in a thin market. It's good that you're moving to 8-20 UTC — that's where the real traffic is.

Now about the sliding activation (TrallingStart): 250 points is quite a combat option. The higher the start, the less often the ADVISER will pull up the stop, but it gives the price a chance to "catch its breath." Experiment boldly.

And what does DeepSeek get for analysis?

Almost all the details of the market, but without SMS and correspondence. To put it quite simply, we throw off such a JSON "lunch" to the AI:

• Trigger: why the neural network was pulled right now (closing the M15 candle or a sharp spike).

• Account: equity and balance — so that the AI understands how alive we are.

• Positions: what is already open, with entry prices, SL/TP and current profit.

• Indicators: classic "borscht set" — ATR, ADX, RSI, CCI, WPR, MFI, price position inside the Bollinger bands and percentage %B.

• The last 30 M15 candles: OHLC and time — so that the neural network sees the picture, and does not point its finger at the sky.

• Feedback history: a few recent AI predictions and whether they turned out to be correct, such as "look, here you are fucked up, and here you are handsome."

In short, the AI gets everything it needs to make a decision, except for the horoscope for today (although... if you add it, maybe the accuracy will increase?).

Regarding the time change to 8-20 UTC, I highly recommend it; night spikes are usually false, and the AI can devour them as a truthful signal. And with a 250—dollar trailing, it's safe, but if the trend is long, you'll take off the fat.
 
Here are the results from this week with version 1.4, Vol=0.01
Screenshot.jpg
I hope this success continues next week. :)
 
Last edited:
DeepSeek, like a real star, decided to rebrand itself — and our good old deepseek-chat model went on a well-deserved vacation, and it was replaced by deepseek-v4-pro and deepseek-v4-flash.

The adviser, not understanding the humor, tried to reach the old model and received an HTTP 400 response like "guy, where are you calling, there is no such model anymore."

What has changed:

Changed the model name in the code from deepseek-chat to deepseek-v4-pro.

Now the AI is back in touch, smarter than before and ready to predict gold movements, rather than drawing errors in logs.

If you have an older version, just download the updated version. Now the advisor is back online, and DeepSeek has stopped pretending to be unavailable. Profit for everyone and no 400 mistakes!
 

Attachments

NPNavigator said:
Hello, colleagues!

Now our "golden" AI advisor has stopped being stubborn as a sheep and has learned to obey his master.

What was added:

• Two new fields in the settings: LLM_BuyThreshold and LLM_SellThreshold

(search in the "AI Settings (DeepSeek)" group).

Previously, the thresholds were tightly sewn in:

prediction >= 0.65 → BUY, prediction <= 0.35 → SELL.

Now you decide for yourself with what confidence the AI should open a deal.

If you want, tighten the nuts to 0.8 and 0.2 in order to trade only "reinforced concrete" signals.

If you want, loosen to 0.6 and 0.4 and catch more movements (but with popcorn, because it can be stormy).

Now, instead of "the robot thinks you're a sucker and ignores your wishes," we have a "henpecked robot" - turn the sensitivity knob and watch as it obediently runs to open positions.

Try it, experiment, and share the results. Version 1.4 is already in the topic header. Profit for everyone!
Click to expand...
hello
i am facing "HTTP 400" error, removed EA from the chart and added again, but still same issue. please give me solution
 
Otabek said:
hello
i am facing "HTTP 400" error, removed EA from the chart and added again, but still same issue. please give me solution
Click to expand...
The HTTP 400 error is not a glitch, it's DeepSeek that has granted an "amnesty" to old models and retired deepseek—chat. The counselor knocks on her door out of habit, and she says, "There's no such model anymore, come back tomorrow."

The full detailed analysis of this problem lies in message No. 62 — there I tell you what to replace the model with so that everything starts up again.

So scroll a little higher, read post No. 62, download the updated version — and let the gold work for us again, not on the log screen!
 
You must log in or register to reply here.
Back
Top