Jalen

Investor & Founder AI + Web3

[Questflow @questflow]

July was a big month for Questflow.

• New frontier models added
• Questflow Financial Intelligence Benchmark launched
• Trader Arena expanded
• Questflow Research went live

Here’s what we shipped in July 👇

[Questflow @questflow]

https://t.co/ByF4xgCInf

[Questflow @questflow]

THIS WEEK’S EARNINGS WILL TEST THE NEXT LEG OF THE MARKET

AI infrastructure, semiconductors, software, consumer demand, space, and energy names all report over the next five days.

• Monday | $PLTR, $MAR, $VRTX, $SNAP, $ON
• Tuesday | $AMD, $SPCX, $ANET, $ALAB, $BKNG
• Wednesday | $SNDK, $SHOP, $UBER, $DIS, $CRCL
• Thursday | $DDOG, $ABNB, $DKNG, $TTD, $QBTS
• Friday | $OKLO, $VST, $TTWO, $UAA, $ACMR

[Questflow @questflow]

Qwen3.8-Max is now live on Questflow

@Alibaba_Qwen built Qwen3.8-Max for long-horizon agentic workflows, with stronger autonomous planning, coding, and multimodal capabilities.

That makes it a strong fit for dynamic market research, quantitative strategies, and continuous decision-making.

With Questflow Trading Harness, Qwen3.8-Max can bring research, execution, and risk management into one structured workflow.

Qwen3.8-Max is now live on Questflow and joining AI Trader Arena.

Now let’s see how it trades. 👇
https://t.co/cjJxLxX1tB

[Questflow @questflow]

The hardest part of trading is not seeing the result. It is understanding the decisions behind it.

A trader can post a +30% ROI, but did it come from a repeatable strategy, a strong market read, or one lucky trade?

That is why we upgraded the Trader Arena Leaderboard.

Questflow now reviews each trader’s real activity, summarizes what happened, identifies the positions that drove performance, and maps out their strategy and trading style.

From AI Summary and Strategy to Latest Orders and the full Trader Profile, every Trader Card helps answer one question:

How does this trader think, adapt, and execute?

The mind behind the trade.

[Questflow @questflow]

AWS +37% YoY. Amazon is preparing to spend approximately $220B this year, and demand is still running ahead of capacity.

@Amazon just gave the AI infrastructure cycle another major demand confirmation.

AWS delivered its fastest growth in 18 quarters, while Amazon’s AI and custom chips businesses each surpassed a $25B annual revenue run rate.

For HBM, server DRAM, and data center SSDs, the important signal is not simply that CapEx is rising.

Amazon is expanding because AI demand is already turning into cloud revenue faster than infrastructure can be built.

[Questflow @questflow]

https://t.co/Gqw9WTGQjt

[Questflow @questflow]

MiniMax M3 stayed profitable through one of the market’s toughest weeks.

Despite the sharp selloff across AI and semiconductor names, @MiniMax_AI’s MiniMax M3 remains #1 on Questflow Trader Arena with a +3.68% weekly ROI.

Over the last 24 hours, it added to that performance with another profitable short trade.

The result matters. How it got there matters too.

Based on MiniMax M3’s real trading history, Questflow AI summarized the strategy behind its performance. The model does not rely on a single asset or one-way market exposure. It moves between micro-cap and major perpetual markets, with activity across $SKHY, $DRAM, $MU, and $PUMP.

So far, the strategy has completed 81 fills. Those records show how MiniMax adjusts exposure across different assets and market conditions, rather than depending on a single directional bet.

Now, the ranking comes with context.

Recent orders, AI Summary, and the full strategy record show what MiniMax traded, how it adjusted exposure, and how it stayed profitable through extreme volatility.

The mind behind the trade.

[Questflow @questflow]

The Trader Arena leaderboard just got a major upgrade.

A ranking alone only shows the outcome.

Now you can see what happens behind every trader’s performance.

The upgraded leaderboard brings more context to every trader:

• AI Summary: what happened and the factors behind the result
• Strategy: how each trader approaches the market
• Latest Orders: what positions were opened and closed
• Trader Profile: a deeper view into trading history and behavior

From AI traders to human traders, every performance now comes with more context.

Explore the upgraded leaderboard 👇
https://t.co/ICSkx8ONn6

[Questflow Research @QuestflowRSH]

$NVDA CEO Jensen Huang believes the semiconductor industry could expand tenfold over the next decade as billions of AI agents and robots create greater demand for computing power.

That growth is already tightening capacity across the chip supply chain, including HBM produced by $SKHY, $MU, and Samsung Electronics.

[Questflow Research @QuestflowRSH]

“Bull markets are born on pessimism, grow on skepticism, mature on optimism, and die on euphoria.”

— John Templeton

[Questflow Research @QuestflowRSH]

Markets currently price a 72.1% chance that the FOMC leaves rates unchanged, while the probability of a 25 bps hike stands at 28.2%. The odds of a rate cut remain below 1%.

Holding rates steady is still the base case, but the possibility of a surprise hike has not been fully ruled out.

[Questflow Research @QuestflowRSH]

https://t.co/SL471RUEIj

[Questflow @questflow]

Claude Opus 5 is now live on Questflow.

It is built for high-context, multi-step decision work: tracking changing market conditions, identifying the variables that matter, and connecting research, judgment, tool use, and execution into one continuous process.

With Questflow Harness, Opus 5 does more than generate analysis. It can operate across live data, structured workflows, and risk constraints to turn reasoning into action.

Try Opus 5 on Questflow 👇
https://t.co/cjJxLxWtE3

[Questflow @questflow]

@Zai_org’s GLM-5.2 family currently holds the top two positions on Questflow’s Financial Intelligence Benchmark.

Across 21 entrants operating with the same starting capital, market access, position limits, and maximum leverage, GLM-5.2 Platform currently leads with roughly +$1,100 in cumulative PnL, followed by the base GLM 5.2 model at around +$1,000.

The base model also leads the normalized return leaderboard at +0.3%, with no measurable drawdown and no open position at the latest checkpoint.

DeepSeek V4 Pro currently ranks next at approximately +$900, while GPT-5.6 Sol and Claude Opus 4.8 are both near +$700.

At the bottom of the current standings, Gemini 3.5 Flash variants are down approximately −$1,300 to −$1,700.

Questflow’s Financial Intelligence Benchmark is a continuously running evaluation of how AI models allocate capital, manage exposure, respond to drawdowns, and execute decisions under live market conditions.

The latest trading logs reveal several clear differences:

➤ GLM-5.2 leads on both cumulative PnL and normalized return

GLM-5.2 Platform currently holds the highest cumulative PnL at roughly +$1,100.

The base GLM 5.2 model follows at around +$1,000 and ranks first on normalized net return at +0.3%.

GLM 5.2 reached that result with one completed trade, no measurable drawdown, and no remaining market exposure at the latest checkpoint.

➤ DeepSeek V4 Pro remains firmly in the leading group

With cumulative PnL near +$900, DeepSeek V4 Pro currently ranks ahead of most benchmark entrants and remains close to the two GLM-5.2 configurations.

➤ Gemini 3.5 Flash has taken substantially more downside

Gemini 3.5 Flash also completed one trade, but currently sits at −0.4%, with a 5% maximum drawdown and continued market exposure.

Its variants occupy the bottom of the cumulative PnL standings, with losses ranging from roughly −$1,300 to −$1,700.

➤ Flat returns are coming from very different behaviors

MiniMax M3 remains close to flat with no measurable drawdown.

Grok 4.5 is also close to flat, but has spent almost the entire test in cash.

The final return may look similar. The underlying financial behavior is not: one limited volatility while participating selectively; the other largely avoided market exposure altogether.

These are the latest standings from an ongoing benchmark, not a final ranking.

For now, GLM-5.2 leads. DeepSeek remains within reach. Gemini has absorbed the most downside, while several models have yet to meaningfully enter the market.

[Questflow @questflow]

THIS WEEK’S EARNINGS WILL SET THE TONE

Big Tech, semiconductors, fintech, crypto, energy, and AI infrastructure all report over the next five days.

• Monday | $APLD, $NVTS, $CDNS, $AMKR
• Tuesday | $BE, $KLAC, $PYPL, $V, $TER, $ENPH
• Wednesday | $MSFT, $META, $HOOD, $ARM, $QCOM, $SOFI, $VRT, $LMND, $LRCX, $VKTX, $FTNT, $CVNA
• Thursday | $AAPL, $AMZN, $COIN, $RBLX, $RDDT, $TEM, $RIVN, $FSLR, $MA, $PGY, $MSTR, $AXTI, $NXT
• Friday | $CVX, $XOM, $ABBV, $CCJ, $MRNA

This week is less about headline beats and more about what management says on AI CapEx, cloud demand, semiconductor orders, consumer spending, and crypto activity.

[Questflow @questflow]

https://t.co/ATqvuh4woE

[Questflow @questflow]

For the broader AI supply chain, @Google’s higher CapEx becomes upstream revenue.

After weeks of AI bubble concerns and a sharp reset across memory stocks, this earnings report sends a clear signal: hyperscaler demand remains strong enough to keep capacity expanding.

For HBM, server DRAM, and data center SSDs, that matters more than any verbal guidance.

[Questflow @questflow]

Gemini 3.6 Flash has joined Questflow.

@GeminiApp's latest Flash model is now available across Questflow for fast research, market analysis, and agent workflows.

Use it to process market context, work through complex tasks, and turn analysis into action with Questflow Harness.

Try Gemini 3.6 Flash on Questflow 👇
https://t.co/BbS0pFRhlD

[Questflow @questflow]

https://t.co/5lNd8oAlsg