FOOTBALL AND BANTER ⚽🥅 @Footbanta

@heisJesse_ Sounds good

The biggest gap between retail and institutional investors may not be information, it may be what they do with the same information they have. Take an 𝗡𝗩𝗜𝗗𝗜𝗔 𝗲𝗮𝗿𝗻𝗶𝗻𝗴𝘀 𝗿𝗲𝗽𝗼𝗿𝘁 for example; Assuming two people open the exact same filing, one of them looks at revenue growth and net income, sees that both went up, then closes the report and checks what other people are saying. then the other starts breaking the numbers he saw down. 𝗙𝗼𝗿 𝗲𝘅𝗮𝗺𝗽𝗹𝗲, NVIDIA’s Hyperscale and ACIE businesses were similar in size, but their quarter-over-quarter growth was very different: roughly 12% vs 31%. both businesses grew, but the second level of analysis asks about knowing where the incremental growth is actually coming from. That can tell you much more about how the business is changing than simply knowing that total revenue increased. ▫▪ And this is why I think the phrase '𝙞𝙣𝙨𝙩𝙞𝙩𝙪𝙩𝙞𝙤𝙣𝙨 𝙬𝙞𝙣 𝙗𝙚𝙘𝙖𝙪𝙨𝙚 𝙩𝙝𝙚𝙮 𝙝𝙖𝙫𝙚 𝙗𝙚𝙩𝙩𝙚𝙧 𝙞𝙣𝙛𝙤𝙧𝙢𝙖𝙩𝙞𝙤𝙣' explanation is becoming less convincing. Because a huge amount of financial information is already public, the difference is often in how deeply it gets processed. A 200-page report is not just two numbers There are at least three things worth doing with a financial report., and they are - ▶ 𝗕𝗿𝗲𝗮𝗸 𝗱𝗼𝘄𝗻 𝘁𝗵𝗲 𝗴𝗿𝗼𝘄𝘁𝗵 Revenue going up is only the beginning. It is good to find out; which segment grew? which customers drove it? which region? is the growth accelerating or slowing? Those details can completely change how you understand the headline number. ▶ 𝗥𝗲𝗮𝗱 𝘁𝗵𝗲 𝗳𝗼𝗼𝘁𝗻𝗼𝘁𝗲𝘀 Micron, for example, disclosed roughly $100B of remaining performance obligations tied to a strategic customer agreement. it is tempting to see $100B and immediately assume that the future revenue is basically guaranteed. But that is not how financial analysis works. You still need to know about; the current revenue, the performance requirements, how flexible the prices are, what will happen if demand or supply changes, The headline number is only the starting point. ▶ 𝗖𝗵𝗲𝗰𝗸 𝘄𝗵𝗲𝘁𝗵𝗲𝗿 𝘁𝗵𝗲 𝗮𝗰𝗰𝗼𝘂𝗻𝘁𝗶𝗻𝗴 𝗯𝗮𝘀𝗶𝘀 𝗰𝗵𝗮𝗻𝗴𝗲𝗱 Sometimes a company moves a business into discontinued operations. The numbers are not necessarily wrong, but the reporting basis has changed. and this means that historical profit may no longer be directly comparable with today's continuing operations. A proper analysis may require reconciling GAAP, Adjusted and Other figures before deciding what the historical business actually earned. in one example, that process produced about $17.8M of continuing-operations earnings, but that number still should not automatically become a future forecast. The calculation aspect matters, and knowing how to use the calculation matters more. ▫▪ This is where 𝗳𝗶𝗻𝗮𝗻𝗰𝗶𝗮𝗹 𝗔𝗜 gets more and more interesting. Historically, institutions used teams of analysts to do this work. And during earnings season, one of the most painful jobs is updating financial models. because a model can contain thousands of formulas across multiple sheets. Actual earnings come out, and someone has to replace forecasts with actual figures, preserve the existing formulas, update cross-sheet dependencies, check that everything still reconciles, find circular references, update charts and summaries Ling-3.0-flash-Fin has been tested on this kind of workflow. In one official demonstration, it updated a Google Q2 2026 financial model by connecting reported figures to the existing spreadsheet structure, checking formulas and updating the summary outputs. the important part is not getting a pile of numbers back, it is producing a model that can still be checked, edited and used. Although there are still limits. AI should not be making the final investment decision for you. Key assumptions, valuations and conclusions still need human review. and that is actually how I think it should work. AI handles the heavy information-processing work, while humans make the judgment. The information gap keeps getting smaller, the ability to process that information properly is becoming the bigger advantage. Try Ling-3.0-flash-Fin on OpenRouter 👇🏾 openrouter.ai/inclusionai/li…

王大锤 @boniusex

2 weeks ago

散户和机构的差距根本原因竟然在这里!! x.com/i/article/2095…

175 96 109 21K 14

@heisJesse_ Let's build

Can this AI update an Excel financial model without breaking everything around it? this was the question behind a recent test of Ling-3.0-flash-Fin. the setup was simple; Take Alphabet’s official Q2 2025 financial report, build a small Excel model with 14 estimated values, and ask the model to replace those estimates with the actual reported figures. But there was one important condition for the AI to take note of, and it was that the original formulas and charts had to stay intact. This was not just about putting numbers into empty cells. The workflow had to read the financial report, identify the correct actual values, update the table, preserve the existing spreadsheet logic, and return a usable Excel file. 𝘁𝗵𝗲𝗻 𝗵𝗼𝘄 𝗱𝗶𝗱 𝗶𝘁 𝗹𝗮𝘁𝗲𝗿 𝗽𝗹𝗮𝘆 𝗼𝘂𝘁? The test was run 5 times using the same configuration. All 5 passed the validation checks and table recalculations. the median runtime was about 42 seconds, with individual runs taking between 25 and 58 seconds. The researchers also kept a separate answer sheet that the model never saw. A program checked 22 key values and calculations, then the resulting Excel files were opened and recalculated to verify the formulas and charts. All five outputs passed. But this does not mean that the model will work perfectly on every financial model. it only shows that, under this specific setup, the process was repeatable. and that difference is very important. The workflow is more interesting than the number, because the first two attempts actually failed, one output became too long and was truncated. yet another did not pass all the required tool parameters, so the file could not be handed over. and after adjusting the configuration, the final five runs worked. That tells me something important about using AI for financial work, and it is that the model does not have to do everything itself. A better setup is from financial report ➡ Ling-3.0-flash-Fin ➡ tools write the spreadsheet ➡ program checks the result ➡ human reviews it The model handles understanding the financial information, the tools handle the spreadsheet operations. and the validation layer checks whether the result is actually correct. That is much closer to a useful financial Agent than simply asking a chatbot to 'update this Excel file.' the test used OpenRouter's free routing, so the model cost shown for these runs was $0. Of course, building the workflow, running the tools and checking the files still takes time and resources. The next question is the one that will matter the most, and it is knowing whether the same workflow can handle a completely different financial model or task and still return a file that is actually usable. that is where the main test begins. Try Ling-3.0-flash-Fin on OpenRouter 👇🏾 openrouter.ai/inclusionai/li… View the open-source model on Hugging Face 👇🏾 huggingface.co/inclusionAI/Li…

Soran @Soranlan

2 weeks ago

让 AI 更新财务模型，数字填对了，原来的公式和图表还在吗？ 这次给 Ling 3.0 Flash Fin 搭了个小工作流：读 Alphabet 财报，把表里 14 个预估值换成实际值，再交回 Excel。 固定配置跑了 5 次，都通过了检查和表格重算，中位耗时 42 秒。 下面是输入表和结果文件的回放。

158 99 105 11K 11

@ToriaIsNerdy @FastXNetwork Good knowledge

Prediction markets are only one piece of the puzzle. The real deal is infrastructure that can turn market data into intelligence for Web3 applications. That’s where @FastXNetwork could connect prediction markets, AI and trading workflows. Learn more at fastxnetwork.com

@MR_0FFICIALL 6/6

You’re making a few USDT transfers every day. Nothing unusual just regular on-chain transactions. Now imagine that, instead of paying around 2.3 USDT in fees for each transfer, you could make the same transfer using just 0.5 USDD through GasFree. That difference may look small at first. But over hundreds of transactions, it starts to add up. 🧮 𝗟𝗲𝘁’𝘀 𝗗𝗼 𝗧𝗵𝗲 𝗠𝗮𝘁𝗵 If an active TRON account averages 2.6 transactions per day: ➜ Typical transfer: ~2.3 USDT ➜ USDD on GasFree: 0.5 USDD ➜ Potential saving: 1.8 USDD per transfer Over one year: 2.6 × 365 × 1.8 = 1,708+ USDD That means the difference in transaction fees could potentially add up to 1,708+ USDD annually at that transaction frequency. 𝗪𝗵𝘆 𝗨𝗦𝗗𝗗 𝗼𝗻 𝗚𝗮𝘀𝗙𝗿𝗲𝗲? ➜ No separate TRX needed for eligible transfers GasFree abstracts the network gas requirement, allowing the applicable fee to be paid in the transferred token. ➜ 0.5 USDD per transfer A simple fee structure makes the cost easier to understand before sending. ➜ Small savings can add up The difference on one transaction may look small. Multiply it across hundreds of transfers, and it becomes much more meaningful. Same transfers. Lower fee friction. More money stays in your wallet. 💚 𝗛𝗼𝘄 𝗧𝗼 𝗚𝗲𝘁 𝗦𝘁𝗮𝗿𝘁𝗲𝗱 𝗪𝗶𝘁𝗵 𝗚𝗮𝘀𝗙𝗿𝗲𝗲 1️⃣ Open a supported wallet Use a wallet that supports GasFree. 2️⃣ Create or access your GasFree wallet Follow the wallet’s GasFree setup process and securely back up your wallet credentials. 3️⃣ Fund your wallet Send supported assets such as USDD to your GasFree wallet. 4️⃣ Choose Send Enter the recipient's TRON address and the amount you want to transfer. 5️⃣ Review the fee Check the applicable GasFree fee before confirming. 6️⃣ Confirm the transaction Approve the transfer in your wallet and let GasFree handle the gas mechanism. That’s the whole idea: You hold USDD → you send USDD → the fee is handled in USDD → no separate TRX balance for eligible transfers. 🔗 Start saving: [USDD × GasFree](usdd.io/gasfree @USDDecentralize @DeFi_JUST @justinsuntron #TRONEcoStar

@Princessubbbb @moovexyz I will give it a try

Have you tried the @moovexyz app yet? I’m curious about the experience from people who have actually used it. What’s the feature you find most useful so far? And more importantly, what do you think could be improved? The best products are built by listening to the people using them every day. Feedback might look small, but those little insights can make the experience much better over time. Would love to hear what the Moove community thinks. 👇

𝐇𝐚𝐬𝐧𝐚𝐢𝐧 @Hasnain007861

11 hours ago

Have you ever tried this app? what’s your favorite feature so far? and what should we improve?

22 2 31 568 1

@Orracklle @Lexacoinbsc 4/4

I came across @Lexacoinbsc recently and, honestly, I almost scrolled past it. Another BSC token, right? But then I started looking at what they are actually building around it, and that is where things got interesting. LexaGroup is building an ecosystem around 👉healthcare, 👉e-commerce, 👉transportation and 👉finance, with LexaCoin positioned as the payment and utility layer. And that healthcare angle matters. WHO says 2.1 billion people faced financial hardship from out-of-pocket healthcare spending in 2022. So the idea of connecting everyday healthcare services with a digital payment layer is not coming out of nowhere. The token runs on BNB Smart Chain with a stated fixed supply of 21M $LEXA. With ✅liquidity locking, ✅a renounced contract, ✅burns and buybacks as part of its token model. But this is where I am keeping my eyes open. ▶️A fixed supply does not create demand. ▶️“Utility” written on a website does not create adoption. And can the wider ecosystem give the token a reason to exist beyond speculation? CA: 0x6fc20e595a8704725DBd160E7c799665706e0bdD

@abbahdigitales @injective @arc Thanks for sharing

I came across this update from @injective and the Arc integration immediately caught my attention Injective is now connected to @arc on Day One, opening the door for users to transfer assets between both ecosystems, trade with native USDC, and launch new assets Exactly what we are looking for Omo 😩 What I find interesting is that this isn’t just about connecting two networks. It creates another path for liquidity, users and builders to move across ecosystems without everything being isolated And the fact that Injective is already teasing more Arc ↔ Injective updates makes this even more interesting For me, the real thing to watch now is what gets built on top of this connection. More liquidity, more applications and more ways for users to interact with both ecosystems could make this integration much more meaningful over time. Day One is only the beginning Curious to see where Injective takes this next. $INJ

@IdaraImeh Interesting

I was watching the CLARITY Act vote yesterday and checking the market at the same time. Then the Senate failed to advance it. Bitcoin was already under pressure, and crypto-linked stocks were moving lower too. What caught my attention wasn't just crypto. It was seeing how one political headline could hit different markets at almost the same time. I had @BingXOfficial open while following the news, so I could switch between stocks and crypto and watch the reaction. It was a good reminder: Sometimes the market story is bigger than the asset you were watching. Were you watching the vote, or did you only notice when the charts turned red? 👀

@MR_0FFICIALL Thanks for sharing

𝐇𝐨𝐰 𝐭𝐨 𝐑𝐞𝐧𝐭 𝐄𝐧𝐞𝐫𝐠𝐲 𝐨𝐧 𝐉𝐮𝐬𝐭𝐋𝐞𝐧𝐝 𝐃𝐀𝐎 If you regularly use TRON for transfers, DeFi, or smart-contract interactions, you don't always need to freeze TRX just to get Energy. JustLend DAO’s Energy Rental gives you another option: rent the Energy you need, for the period you need it, and use it to reduce the TRX cost of eligible TRON transactions. Here’s how the process works 👇 ➠ 1. Open Energy Rental Go to the JustLend Energy Rental page and connect your TRON wallet. ➠ 2. Enter the Energy amount Choose how much Energy you need based on your expected transactions. For example, the interface currently shows 300,000 Energy, estimated to be enough for around 7 transactions. ➠ 3. Choose your rental duration You can select: ▫️ 3 hours ▫️ 1 day ▫️ 3 days ▫️ Custom duration This makes the model more flexible than permanently locking capital for Energy. ➠ 4. Choose the receiving address You can rent Energy to your connected wallet or specify another TRON address. That can be useful when you're preparing transactions for a separate wallet. ➠ 5. Review the prepayment Before confirming, JustLend shows the required prepayment and potential refund. In the example shown, the interface displays 57.62 TRX prepayment, with 24.51 TRX listed as the minimum refund if the rental ends early. ➠ 6. Confirm the rental Review the Energy amount, duration, receiving address and payment details, then connect your wallet to complete the rental. Why this matters The interesting part isn't simply that Energy can be rented. It's the capital-efficiency trade-off. Instead of acquiring Energy by staking a large amount of TRX, users can access Energy temporarily when they actually need it. That can make sense for users who: ▫️ Need Energy for occasional transactions ▫️ Don't want to lock large amounts of TRX ▫️ Want more predictable transaction costs ▫️ Need Energy for another TRON address ▫️ Prefer an on-demand resource model The current JustLend interface also shows 47.84B+ total Energy, with more than 15.87B Energy in rent, alongside 82,385 total renters and 16,771 resource providers. That gives a useful picture of Energy Rental as a two-sided resource market rather than simply another DeFi button. For me, the key takeaway is simple: TRON Energy doesn't always have to mean “stake first, transact later.” With Energy Rental, the approach can be “need Energy → rent it → use it → move on.” [Try JustLend Energy Rental](app.justlend.org/energyRental?l… #TRONEcoStar @justinsuntron @DeFi_JUST

JUST DAO @DeFi_JUST

7 days ago

⚡Energy just got cheaper on #JustLendDAO Base rate cut: 15% → 8% ⛽Current pricing 100,000 Energy = 4.668 TRX (46 sun/Day) Lower Energy cost → cheaper smart contract execution. Efficiency is part of every good DeFi strategy. 🔋Check pricing: app.justlend.org/energy?lang=en…

10 14 43 5K 2

@tashamulti Wow

The RWA market is getting really interesting on Hyperliquid 👀 Between July 13–19, RWA markets did $25.1B in trading volume …that’s about 52% of Hyperliquid’s total weekly volume. So this isn’t just about tokenized assets sitting on-chain anymore. → Stocks and commodities are being actively traded on-chain → HIP-3 is expanding the types of markets builders can create → Ondo’s tokenized assets are also becoming part of this ecosystem And with the Fed decision coming up, I’m curious to see how changes in liquidity and market conditions affect this side of crypto. $HYPE is one of the tokens connected to the Hyperliquid ecosystem. #BingX #HYPE