On Oct. 16, Bitcoin jumped more than 10% in a matter of minutes as false report circulated saying that the US had approved a long-awaited ETF that invests directly in the cryptocurrency. It erased most of that increase just as quickly when BlackRock said its application was still with the SEC.

Bitcoin’s Big Moves Are Being Magnified by a Protracted Lack of Liquidity

As per data from lookonchain, FalconX, a well-known crypto trading platform, seemed to be offloading MKR. Over the last week, FalconX moved a sum of 9,085 MKR, which was worth $12.8 million, into OKX and Binance. At press time, there were 7,261 MKR ...

MKR sees large sell-offs; will traders stay put in the face of this adversity?

Maker Protocol’s token has seen a surge in supply across crypto exchanges. Meanwhile, trading firm FalconX contributed to the increase after sending MKR coins worth around $12.8 million to Binance and OKX within the last seven days.

Maker (MKR) price turns bearish as exchange supply soars

Ripple price rallied to the $0.55 level, after yielding 6.2% weekly gains. As the asset sustains above the $0.50 psychological barrier, its rally is supported by bullish on-chain metrics. That said, there is a possibility of a correction in XRP price, if the asset drops to support at $0.50, it could plummet lower.

Cryptocurrencies Price Prediction: Ripple, Bitcoin & Maker– European Wrap 30 October

FalconX transferred $12.8 million worth of MKR tokens to exchanges over the past week. According to data from Lookonchain that tracks recent transactions, the firm sent the asset to centralized exchanges OKX and Binance, contributing to the rise in the asset’s supply.

Maker price drop continues as selling pressure on MKR brews across exchanges

Crypto investors can earn rewards by pledging Ether tokens to help operate the Ethereum blockchain, a return some cite as a potential prop for the coin. But the staking payout on pledged tokens has fallen to an annualised 3.5%, near the lowest in at least 10 months and well below a recent peak of more than 8%.

Bitcoin’s rival ether falters in shadow of elevated treasury yields

The market on-chain analytics platform Lookonchain has been closely observing FalconX’s trading activities over the past few days. Observing a new withdrawal activity, the market analytics platform reported that FalconX once again withdrew 2.5 million ...

$4M LDO Withdrawal by FalconX Sparks Excitement! Is a Bull Run Coming?

The TRADE speaks to Bhav Trivedi, manager for dealing UK at AustralianSuper, to discuss the interplay and key differences between trade cost analysis (TCA) and liquidity provision analysis (LPA), the key considerations when applying these in trade ...

Asset Classes

While new addresses refrained from transacting, LDO transfers increased. For the second time in three days, FalconX has moved a ton of Lido Finance [LDO] tokens into self-custody. The first time the institutional crypto platform transferred LDO was on 11 ...

Lido accumulation takes an interesting turn with this institution

FalconX

falconx

The abbreviation OTC stands for Over-the-Counter, which means special trading/exchange service for institutional investors, which allows the purchase of investment assets (e.g. stocks) or cryptocurrencies, bypassing the standard exchange/marketplace procedure.  Based on “most popular conception of OTC trading revolves around massive off-market deals, like when companies such as MicroStrategy make multi million-dollar purchases using OTC desks run by Coinbase or Kraken.” Cointelegraph review Main attributes for crypto OTC deals Although, as we said above, an OTC exchange involves an exchange outside of an exchange or exchange stack, from an AML perspective, it's the amount of the transaction that matters - it's usually spelled out on these companies' websites as an exchange above $50,000 dollars. African crypto exchange Yellow Card in its divides two different OTC trading markets: Principal OTC trading (trading with own funds of principal) and Agency OTC Trading (using 3rd party funds). review According to provided research of different OTC desks we may classify them in 3 large groups by their business activity: -OTC desks owned by crypto platforms -Online OTC desks -Offline OTC desks Which jurisdictions large cryptoplatforms use for OTC exchange? has described companies, which it is using for OTC operations: for Canadian and US users it’s FORIS DAX (registered in Canada) and for all other countries - CRO DAX Limited registered on Cayman Islands. Crypto.com In the we have no information which license is used for OTC trading. However in the list of companies in Annex F there is information that one of these licensed isn’t used for OTC. So probably it means that other companies in the list CAN offer OTC desk and 2 main companies which offer any Kraken services to EEA citizens - are registered in BVA. Kraken user agreement According Binance we have no details which entities does it especially use for OTC deals exactly. However, at least it has 1 company having “OTC” in its title - FiveWest OTC Desk (Pty) Limited registered in South Africa. Coinbase ассording to provides OTC services through special platform for institutional investors - Coinbase Exchange (previously Coinbase Prime). Operations are provided by 2 licensed entities - Coinbase Inc (USA) and Coinbase Custody International Limited (Ireland). information on its website Bitstamp exchange has launched in 2022, it has no 3rd party OTC provider with amount of orders between $50,000 - $500,000. Main companies providing services are registered on BVA, Luxembourhg, UK. OTC desk Bybit provides for crypto - USDT trading services (no fiat exchange) with maximum amount - $3,8 mln USDT per 1 transction. These and other services on Bybit website are provided by the company Bybit Fintech Limited registered in the UK in April 2023. Interestingly, on this platform, OTC is simply a sub-species of P2P trading for large sums and the platform acts as an agent for this exchange. The OTC desk on Kucoin and MEXC is organized in a similar way. OTC trading OTC desk - minimal amount for 1 order is $200,000 and it also works through p2p agent scheme. also provides an agent scheme and abilty to trade instantly with partner OTC trader - Bill Brindise. Gate exchange Bitfinex Yellow Card has OTC desks in the following counties: Ghana, Uganda, Cameroon and South Africa. Other OTC providers Also we’ve observed some online platforms promoting themselves as special OTC platforms and operating in the following jurisdictions: ZondaCrypto (Switzerland, Estonia, Poland, Slovakia) Encryptus (Australia, UAE, Estonia ) Bitcoin Reserve (Estonia) Coinspaid (Estonia, Lithuania, Poland) HashKey (Japan, HonkKong) Enigma (United Kingdom) Merkeleon (Austria) Coinsfera (UAE) FalconX (USA, Malta, United Kingdom) CoinMotion (Finland) Independent Reserves (Australia, Singapore) Offline OTC exchange In this article we have gathered more information about OTC transactions that take place online, but they are also conducted offline - for cash or non-cash payment. The most information about legal offline exchanges that conduct transactions for these amounts in such jurisdictions as Hong Kong and UAE. Exchanges operating there legally also position themselves as platforms for OTC transactions. Some OTC exchanges operating in Hong Kong are collected in this list provided on . Bitget's website In the future, it would be interesting to study what percentage of exchanges are not only online OTC in countries where crypto is regulated, but also in countries with weak regulation - e.g. Armenia, Moldova, Russia, Turkey. In many countries of the world there are no clear AML requirements for the exchange of cryptocurrencies. Because of this, gray exchange for large amounts is actively developed in them. This is an important factor that strongly influences the global cryptocurrency market. Licensing of OTC desks Basic approach that in most jurisdictions there is no OTC trading license. This type of activity is regulated as a standard crypto exchange or currency exchange approach. The vast majority of existing OTC desks operate under a VASP license as online platforms and cryptocurrency exchange companies. The fact that transactions are conducted in OTC format does not affect the required type of license, only imposes additional AML requirements. In some cases, they may operate as brokers or offline exchanges (in countries where offline exchangers require a different type of license than VASPs). Conclusion The regulation of OTC exchange and its separation into a separate type of cryptocurrency transactions is relatively recent and is still in the formation stage. Based on our research we may conclude the following facts about OTC exchange: You need a VASP license to provide OTC exchange List of most popular jurisdictions for OTC exchange mostly matches the list of jurisdictions with the best regulation of cryptocurrencies and VASP licensing procedures Agency OTC does not seem to require any special licensing, while standard OTC using the company's own funds necessarily requires a VASP or broker license The same requirements and restrictions apply to the regulation of OTC exchanges as to regular exchanges from an AML perspective. However, in some cases there may be additional compliance requirements as for online exchange of large sums of money In some jurisdictions (e.g. USA) you must have institutional investor status to access OTC exchange. This research is not the truth in the last instance, we have only presented the facts that we managed to establish, if you found any inaccuracies or you have your own opinion - feel free to write me on or . Twitter Telegram

What is OTC in Crypto and Does it Require a Special License?

BENGALURU, India, November 2nd, 2023/Chainwire/-- (IBW), the country’s flagship series of blockchain and Web3 events held in Bangalore from Dec.4-10, is thrilled to announce the speaker lineup for the main IBW Conference. India Blockchain Week The conference’s headline speakers include Polygon Co-Founder Sandeep Nailwal, Aptos Co-Founder Mo Shaikh, Ava Labs Founder & CEO Dr. Emin Gün Sirer, TRON Co-Founder Justin Sun, and The Sandbox Co-Founder Sebastien Borget. Additional notable speakers will be Rene Reinsberg (Co-Founder of Celo) Prabhakar Reddy (Co-Founder of FalconX), Aniket Jindal (Co-Founder of Biconomy), Zhen Yu Young (Co-Founder of Web3Auth), Irene Wu (Head of Strategy at LayerZero), Manish Agarwal (CEO at IndiGG), Dr. Kang Li (Chief Security Officer at Certik), Simon Kim (CEO & Managing Partner at Hashed), Kelvin Koh (Co-Founder & Managing Partner of Spartan) among many others. Taking place at the Sheraton Grand in Bangalore on December 6-7, the IBW’23 Conference will have 100+ innovators, entrepreneurs, decision-makers and thought leaders across the entire spectrum of the blockchain industry engaged in keynote speeches, panel discussions, fireside chats, and moderated debates. For example: Sandeep Nailwal will engage in a fireside chat, discussing the current progress and future of Polygon 2.0, and sharing insights about the Indian Web3 ecosystem Mo Shaikh will deliver a keynote speech focused on the Move ecosystem and how Aptos enhances scalability and speed to enhance user experience Dr. Emin Gün Sirer will participate in a fireside chat, discussing strategies for fostering Web3 adoption through subnets and Avalanche's India strategy. Justin Sun will have a virtual fireside chat, delving into general crypto trends and offering perspectives on global regulation, stablecoins, and the Tron ecosystem. Sebastien Borget, in his keynote speech, will explore the future of the Metaverse and how brands can unlock value in digital worlds and their massive plans for India. “India is taking full advantage of Web3 technology across industries like social media, entertainment, commerce, and open financial systems”, said Mo Shaikh, Co-Founder & CEO of Aptos Labs. “The Aptos India ecosystem, growth and community is already expanding on a daily basis faster than most others. Our global enterprise partners are taking notice of the blend of market maturity and innovation happening across the country. We are dedicated to the expansion of Web3 in India and look forward to Aptos being a catalyst for the decentralized movement.” Besides the two-day headline conference, the week-long Web3 festival will feature other prime Web3 events such as ETHIndia, Polygon Connect, Polkadot Pulse, FILBangalore and many more. This is a unique opportunity to engage with emerging talent, attract new developers to the blockchain arena, and expand the blockchain ecosystem in India. "We're launching India Blockchain Week (IBW) because we see the synergy between India and Blockchain as not just a game-changer, but our collective destiny," said Tak Lee, CEO of Hashed Emergent, host of the India Blockchain Week (IBW) Conference. About India Blockchain Week (IBW) India Blockchain Week (IBW) is a series of blockchain and web3 events, spanning from 4-10 December 2023 across Bangalore, India. IBW is a week-long array of 50+ web3 events, including flagship events such as the headline IBW Conference, ETHIndia, Polygon Connect, Polkadot Pulse, FILBangalore and many other side-events, networking events, workshops, parties etc hosted by Indian & global blockchain industry stakeholders. The IBW Conference is the two-day headline event anchoring India Blockchain Week (IBW) on 6-7 December 2023 at the Sheraton Grand, Bangalore and hosted by Hashed Emergent - a team and brand dedicated to accelerating the mass adoption of Web3 in India. For more information about IBW, visit: | | | Website Twitter LinkedIn Telegram Contact Irshad Ahmed gm@indiablockchainweek.com This story was distributed as a release by Chainwire under HackerNoon’s Business Blogging Program. Learn more about the program . here

Sandeep Nailwal, Mo Shaikh, Dr. Emin And Justin Sun To Speak At India Blockchain Week

সন্দীপ নেইলওয়াল, মো শেখ, ডাঃ এমিন এবং জাস্টিন সান ইন্ডিয়া ব্লকচেইন সপ্তাহে বক্তৃতা করবেন

Sandeep Nailwal, Mo Shaikh, Dr. Emin und Justin Sun sprechen auf der India Blockchain Week

Sandeep Nailwal, Mo Shaikh, Dr. Emin y Justin Sun hablarán en la India Blockchain Week

Sandeep Nailwal, Mo Shaikh, le Dr Emin et Justin Sun prendront la parole à la semaine indienne de la blockchain

संदीप नेलवाल, मो शेख, डॉ. एमिन और जस्टिन सन इंडिया ब्लॉकचेन वीक में बोलेंगे

サンディープ・ネイルワル氏、モー・シャイク氏、エミン博士、ジャスティン・サン氏がインドブロックチェーンウィークで講演

Sandeep Nailwal, Mo Shaikh, Dr. Emin 및 Justin Sun이 인도 블록체인 주간에서 연설합니다

Sandeep Nailwal, Mo Shaikh, Dr. Emin e Justin Sun falarão na India Blockchain Week

Сандип Наилвал, Мо Шейх, доктор Эмин и Джастин Сан выступят на Неделе блокчейна в Индии

Sandeep Nailwal, Mo Shaikh, Dr. Emin ve Justin Sun Hindistan Blockchain Haftasında Konuşma Yapacak

Sandeep Nailwal, Mo Shaikh, Tiến sĩ Emin và Justin Sun sẽ phát biểu tại Tuần lễ Blockchain Ấn Độ

Sandeep Nailwal、Mo Shaikh、Emin 博士和孙宇晨将在印度区块链周上发表演讲

Hi Deng Chao, welcome Please tell us about your journey and the motivation behind ? Ishan Pandey: to our series “Behind the Startup.” HashKey Capital Hello, Ishan. Thanks for the interview invitation. This is DC, CEO of Hash Key Capital and we are an asset manager investing in blockchain technology and digital assets, managing over US$1 billion in client assets since inception. Our goal as one of Asia's largest crypto funds and an early corporate investor in Ethereum is to help bring cryptocurrency into the mainstream while also bridging the gap between the Web2 and Web3 economies. Since its inception in 2015, Hong Kong- and Singapore-based venture capital firm Hash Key Capital has invested in over 500 projects across a wide range of sectors, from Layer1 protocols and crypto finance to Web3 infrastructure, NFTs and the Metaverse. Deng Chao: exclusively Besides Ethereum, our portfolio also includes Cosmos, FTX, Coinlist, Aztec, Blockdaemon, dYdX, imToken, Animoca Brands, Falcon X, Space and time, Mask Network, Polkadot, Moonbeam and Galxe (Originally Project Galaxy). : Vested Interest Disclosure The author is an independent contributor publishing via our . Be it through direct compensation, media partnerships, or networking. The author has a vested interest in the company/ies mentioned in this story. HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR brand-as-author program Tell us about the License that you obtained from the Securities and Futures Commission (‘SFC’) of Hong Kong. Why is this license important and how will it help the company? Ishan Pandey: Since 9th Sep 2022, HashKey Capital Limited has secured SFC’s Type 9 license uplift to manage virtual asset portfolio. HashKey Capital Limited will be among the first group of licenced virtual asset managers in Hong Kong, and will be able to manage funds that invest solely in virtual assets in accordance with SFC regulations. Obtaining a Type 9 licence demonstrates our continued dedication to developing Hong Kong's and Asia's blockchain ecosystems. Deng Chao: 100% What are the current business lines of Hash Key Capital and how does it help the crypto industry as a whole? Ishan Pandey: o: Hash Key Capital is an asset manager with three dedicated funds investing exclusively in blockchain technology and digital assets. We see ourselves as a Web3 ecosystem builder, not just following a thematic-driven investment strategy. Deng Cha Besides that, we focus on post-investment management, including empowerment that helps projects expand their markets, design products, find target audiences and provide strategic support through active management. Meanwhile, we have been actively engaged in ecosystem contribution. The HashKey ecosystem is the official partner with Polkadot and Dfinity while helping them host high-quality Bootcamp and hackathons. What are your views on upcoming DeFi and Stablecoin regulations by US regulators? Ishan Pandey: Devcon 6 has shown that regulatory changes in DeFi and stablecoin will be significant in the future. The industry’s DeFi players have already begun to embrace compliance, especially those large companies. Our opinion is that stablecoins such as USDC will go further in the future. Overcollateralization will be the new norm for most stablecoins. For many projects, compliance will also become a critical factor that needs to be concerned about. Although DeFi is globally distributed, most of the teams behind DeFi will not go against regulators. The upside of more regulation is that scrutiny will make it easier for institutions to come in. Deng Chao: Do you think the declining Bitcoin and the insolvency of big cryptocurrency players like Celsius will adversely affect investments in the blockchain industry? Ishan Pandey: By the nature of competition, it is quite normal that players that start to lag behind will be washed out in the early stages of the market. When the market condition is good, it is hard to tell which projects are doing better. Once it goes down, it actually helps us automatically screen projects. Deng Chao: Do you think regulations and compliance are necessary to increase trust and sustain growth of the blockchain industry? Ishan Pandey: Regulation and compliance are indeed essential for traditional investors to secure their trust in the industry. In our opinion, it depends on the institutions and what they are looking for. Meanwhile, adding many traditional users to the market, that is to say, mass adoption, will certainly require compliance and regulation. Hash Key Capital has insisted on being fully compliant since the beginning of our business, and we are also in constant contact with the regulators to comprehend the visionaries from the authorities' perspective. Despite our Type 9 license uplift in Hong Kong, we are also in the process of applying for a CMS license in Singapore. Deng Chao: According to you, what is the next big thing in Blockchain technology? Ishan Pandey: There are two things to consider here. First, Ethereum will become more and more user-friendly from a technical standpoint. With roll-up and sharding running at total capacity, Ethereum will become an all-around better platform. Second, blockchain needs to bring in more traditional users, and it is vital to offer products and services that appeal to them. In this regard, the recent discussion of abstract accounts can be considered a phenomenon. Deng Chao: Don’t forget to like and share the story!

Regulatory Changes in DeFi and Stablecoin will be Significant 

Los cambios regulatorios en DeFi y Stablecoin serán significativos

DeFi और Stablecoin में नियामक परिवर्तन महत्वपूर्ण होंगे

In 2021, the number of mobile app downloads reached 230 billion. The very thing is – it’ll keep growing. Can you wrap your mind around that for a second? Along with that, what is it you see right before going to sleep and first thing in the morning? My guess is a mobile screen: notifications, alerts, messages, anything. Right, we do the same. As an integral part of the tech junkie community, we’re always on the lookout for apps that make a difference. This time, we’ll tell you about some of the 10 most promising start-ups that are about to roll out their mobile apps in 2022. Get ready. 1. Weavit’ Weavit’s co-founder Emmanuel Lefort thought up a way to make note-making even more special. The basically mimics how the human brain entertains ideas and gives them structure. new app by Weavit It works like this: users dictate or type in their thoughts and ideas, and Weavit’s NLP-powered database links those notions to certain events, locations, people, and topics that are out there in the online universe. Here’s an example. You may think of introducing your idea to some individuals at an upcoming event. You’d type something along the lines of “say this and this at the ABC event, and make sure Chris T. and Hannah K. hear it.” Now, Weavit’s app identifies both the event and people’s names and offers you other event-related content. It can even display the mentioned persons’ companies or past meetings. And at this point, you don’t have to lift a finger – the system is automatic. Weavit founders made sure that all the data involved is encrypted. They call the app a “Shazam for your thoughts.” That’s a fact – you only need to let those thoughts free, and the app will do the organization. 2. Niantic Founded by John Hanke, Niantic delivers innovative software solutions and is most known for building such augmented reality apps as Ingress and Pokemon Go (recall the hype?). But that’s not what we’re here for. The company has recently received to create a real-world metaverse. $300 million in funding from a worldwide technology investor Coatue What are we dealing with? While many tech companies focus on metaverses, Niantic decided to go a step further. The startup revealed their plans to launch a universe that belongs in the real world – that is, with AR-powered technology, people can go out in the open and explore their surroundings using Niantic’s tools. These same tools activate actual objects, allowing people to enjoy an enriched reality while still connecting with their peers. The most value here lies in the fact that Niantic strives to bring people together instead of locking them inside and tying them to devices. , “At Niantic, we believe humans are the happiest when their virtual world leads them to a physical one.” Hanke himself stated Meet you out there. 3. nate The founder & CEO of nate, Albert Saniger, solved the struggle for many shoppers. Nate's well-funded – we’re talking – new app combines social media interactions and shopping in one. $38 million Here’s the catch: ate’s app saves user data needed for purchasing, such as payment and shipping details. So, when a user intends to buy anything from a supported online platform, they just need to redirect their purchase to the nate app. It handles the rest – and there’s no need to go through a lengthy checkout process every time. Nate’s users can buy products from , which make up 60% of the country’s e-commerce market. 2.1 million online US stores 4. Papaya Patrick Kann launched Papaya keeping in mind that payments can get as easy as ever. Same goes for anything else that comes with it, like scheduling and tracking – with Papaya, users can spend more time on what’s important and let the app do the work. The start-up raised a whopping from Sequoia Capital – a well-known capital group that recently invested in Airbnb, Stripe, and Apple. That’s impressive for a fresh company. $50 million in funding The mechanics: app users take a photo of a bill that needs to be paid (whether it’s healthcare services, rent or mortgage, parking, anything), select their payment method, and Papaya finishes the process for you. It also saves your billing history just to make sure nothing gets missed out. Such an untraditional way of paying bills is loved by many: it reduces stress, saves time, and is universal for different bill types. More benefits include Papaya’s as well as its speed of bill processing. All that’s needed from you is a mobile device and payment details. bank-level security backed by HIPAA and PCI compliance 5. Dream Games Soyner Aydemir managed to create a leading mobile gaming business – Dream Games – that has recently from Index Ventures, Makers Fund, IVP, and other major venture capitals. The start-up doubled its value in only 6 months. Let’s see why. gained $255 million in funding Dream Games’ most popular feature, Royal Match, ended up in the top 20 mobile games in the US, UK, and Germany. The latest funding is expected to develop the game even further. that “Dream Games approaches gaming in a way Pixar approaches filmmaking.” Aydemir stated The key points here are engagement and immersion. Users keep coming back for a rich experience regardless of the competition in the gaming world. As of now, Dream Games has gained around from investors. With more money allocated to mobile games, developers get a chance to push the limit further – and create gaming worlds no one would’ve expected to witness. $468 million 6. Honor Honor, a senior care startup founded by Seth Sternberg, in Series E funding and debt financing. The concept behind Honor was born as Sternberg quit his job at Google to provide care for his elderly mother. As of now, the unicorn startup’s value is over . And, Honor’s family app helps thousands. received $370 million $1.25 billion What does it do? With the Honor Family app, users can choose a date, location, and time to set up care visits. They receive an extensive care note including the completed tasks from the chosen carer, who also oversees other people visiting your loved one. Care professionals track any wellness indicators you’re worried about. An app like this makes a difference in the COVID-affected world. It enables caring for your loved ones from a distance and making sure they’re in good spirits. Regardless of your location, Honor brings family a bit closer. 7. Instacart Apoorva Mehta’s creation, Instacart, has been familiar to the masses for a while. Yet, it’s expected to augment after from investors Andreessen Horowitz, Sequoia Capital, D1 Capital Partners, Fidelity Management & Research Company, and T. Rowe Price Associates. securing $265 million in funding The point? Instacart simplifies grocery shopping by letting customers place orders online and get them delivered right to the door. These days, the app is markedly in demand. Instacart went further than decreasing social traffic in grocery stores. It now delivers prescription drugs, sports equipment, home decor, and practically anything else that helps people handle the pandemic with more convenience. Instacart currently works with . Now that social distancing and delivery services have become a new norm, start-ups like this one keep highlighting the importance of safety without sacrificing comfort. over 600 merchants across 45,000 US and Canadian stores 8. BillEase BillEase in a Series B funding round guided by the growth capital arm of Hubert Burda Media, BurdaPrincipal Investments. Founded by Ritche Weekun and his colleagues, the platform offers buy-now-pay-later services (BNPL) to Filipino residents. received $11 million In detail: app users can purchase goods and pay later with a digital wallet, direct debit, bank transfer, over-the-counter payment at smaller physical stores, or by linking their bank account to the app. BillEase works with 500 sellers and like GCash, PayMaya, Coins.ph, GrabPay and Shopee Pay. offers extra payments for digital wallets BillEase has a few features that make it stand out. It enables clients to create formal credit records and improves its functioning to make credit scores effortless. BillEase also owns its , with which over 90% of BNPL loans can be approved. With the investments aimed at further app development, the Philippines is on its way towards a better credit system. credit, fraud, and payment software stack 9. Masters In a funding round by Sweet Capital, Mucker Capital, Goodwater Capital, and Luxor Capital, Greg Drach’s app Masters . The app gives users an opportunity to work out along with the most known global athletes and get their guidance on training. gathered $2.7 million worth of investment According to , the company has signed up X Games athletes, three-time Olympic snowboarding champion Shaun White; World and nine-time U.S. champion in 3,000-meter steeplechase, Emma Coburn; surfing world champion, Kai Lenny; soccer superstar and Golden Ball winner, Ada Hegerberg; and two-time Wimbledon tennis champion, Petra Kvitova. And the list goes on. Techcrunch All the courses come in a format of high-quality videos, and every course merges thousands of active individuals. Each of them can easily connect to global sports legends with a functional device, and the entire community comes together to become healthier and stronger. Imagine the impact this has! 10. Pluang Pluang, an investment app created by Claudia Kolonas, Richard Chua, and Aditya Cha, closed off a Series B round led by Accel India with in funding. It also received investments from founders of Axie Infinity, Alexa von Tobel (ex-CEO of Learnvest), Sujata Bhatia (Monzo COO), Raghu Yarlagadda (FalconX CEO), The Chainsmokers, BRI Ventures, Gold House, and previous investors – Square Peg, Go-Ventures, UOB Venture Management, and Openspace Ventures. The start-up focuses on giving Indonesians opportunities to grow their wealth through investing. Pluang plans to expand and operate in other regions of South Asia. $55 million gained The point is: Pluang first started with gold and since then expanded the assets list to include indexes, mutual funds, and crypto assets, with plans to invest in US stocks. Every transaction is secured with an appropriate protocol. What’s more, the start-up partners with Indonesian tech companies to broaden their reach and sell more investment products. Pluang has also such as Gojek, DANA, Tokopedia and Bukalapak. become part of Indonesian super apps From January 2020 to November 2021, Pluang witnessed 22x growth in monthly transacting users and 28.5x growth in users with an active balance. As of now, there are over 4 million active users in Indonesia. And it doesn’t seem to stop moving forward. We’ve only covered a tiny fraction of all the promising tech start-ups the world is about to see skyrocket this year. One day, yours might end up on our next list.

10 Promising Mobile Apps by Startups to Watch Out for in 2022

Authors:
(1) Keivan Alizadeh;
(2) Iman Mirzadeh, Major Contribution;
(3) Dmitry Belenko, Major Contribution;
(4) S. Karen Khatamifard;
(5) Minsik Cho;
(6) Carlo C Del Mundo;
(7) Mohammad Rastegari;
(8) Mehrdad Farajtabar. Table of Links Abstract and 1. Introduction 2. Flash Memory & LLM Inference and 2.1 Bandwidth and Energy Constraints 2.2 Read Throughput 3 Load From Flash 3.1 Reducing Data Transfer 3.2 Improving Transfer Throughput with Increased Chunk Sizes 3.3 Optimized Data Management in DRAM 4 Results 4.1 Results for OPT 6.7B Model 4.2 Results for Falcon 7B Model 5 Related Works 6 Conclusion and Discussion, Acknowledgements and References 4.2 Results for Falcon 7B Model To verify that our findings generalize beyond OPT models we also apply the idea of LLM in flash to Falcon model. Since, the base line Falcon model is not sparse, we used a sparsified (relufied) version with almost the same performance as that of the base version (Mirzadeh et al., 2023). Similar to previous section, we present the results obtained under the condition that approximately half of the model size is available for use in DRAM. Predictors. In the Falcon 7B model, predictors of rank r = 256 are used for the initial 28 layers, and r = 1152 for the last four layers. Window Configuration. Our model reserves memory for a window containing the last 4 tokens. This setup utilizes 33% of the Feed Forward Network (FFN). In terms of memory allocation, embeddings take 4.2% of the model size, attention weights account for 19.4%, and predictors require 4%. The active portion of the FFN, given our window size, is 25.3% (calculated as 0.33 × 76.8). Overall, this amounts to 52.93% of the model’s total size. This paper is available on arxiv under CC BY-SA 4.0 DEED license. Authors: (1) Keivan Alizadeh; (2) Iman Mirzadeh, Major Contribution; (3) Dmitry Belenko, Major Contribution; (4) S. Karen Khatamifard; (5) Minsik Cho; (6) Carlo C Del Mundo; (7) Mohammad Rastegari; (8) Mehrdad Farajtabar. Authors: Authors: (1) Keivan Alizadeh; (2) Iman Mirzadeh, Major Contribution; (3) Dmitry Belenko, Major Contribution; (4) S. Karen Khatamifard; (5) Minsik Cho; (6) Carlo C Del Mundo; (7) Mohammad Rastegari; (8) Mehrdad Farajtabar. Table of Links Abstract and 1. Introduction Abstract and 1. Introduction 2. Flash Memory & LLM Inference and 2.1 Bandwidth and Energy Constraints 2. Flash Memory & LLM Inference and 2.1 Bandwidth and Energy Constraints 2.2 Read Throughput 2.2 Read Throughput 3 Load From Flash 3 Load From Flash 3.1 Reducing Data Transfer 3.1 Reducing Data Transfer 3.2 Improving Transfer Throughput with Increased Chunk Sizes 3.2 Improving Transfer Throughput with Increased Chunk Sizes 3.3 Optimized Data Management in DRAM 3.3 Optimized Data Management in DRAM 4 Results 4 Results 4.1 Results for OPT 6.7B Model 4.1 Results for OPT 6.7B Model 4.2 Results for Falcon 7B Model 4.2 Results for Falcon 7B Model 5 Related Works 5 Related Works 6 Conclusion and Discussion, Acknowledgements and References 6 Conclusion and Discussion, Acknowledgements and References 4.2 Results for Falcon 7B Model To verify that our findings generalize beyond OPT models we also apply the idea of LLM in flash to Falcon model. Since, the base line Falcon model is not sparse, we used a sparsified (relufied) version with almost the same performance as that of the base version (Mirzadeh et al., 2023). Similar to previous section, we present the results obtained under the condition that approximately half of the model size is available for use in DRAM. Predictors. In the Falcon 7B model, predictors of rank r = 256 are used for the initial 28 layers, and r = 1152 for the last four layers. Predictors. Window Configuration. Our model reserves memory for a window containing the last 4 tokens. This setup utilizes 33% of the Feed Forward Network (FFN). In terms of memory allocation, embeddings take 4.2% of the model size, attention weights account for 19.4%, and predictors require 4%. The active portion of the FFN, given our window size, is 25.3% (calculated as 0.33 × 76.8). Overall, this amounts to 52.93% of the model’s total size. Window Configuration. This paper is available on arxiv under CC BY-SA 4.0 DEED license. This paper is available on arxiv under CC BY-SA 4.0 DEED license. available on arxiv

Large Language Models on Memory-Constrained Devices Using Flash Memory: Results for Falcon 7B Model

Amazon Falcon Lite vs OpenAI ChatGPT - The Large Language Model Battle

Non-fungible tokens are tokens based on the Ethereum network released to represent some digital piece of art. These tokens are unique in that each of them has special characteristics and cannot be exchanged 1 for 1. The launch of NFTs happened in 2017, and ever since then, the creation world has vastly grown. NFT As digital assets, they aim to create scarcity and digital ownership. Moreover, they provide interoperability across multiple platforms. They are common in the art and games industry. Due to the vast Ethereum growth and wide adoption of the crypto world, with large companies like Tesla investing in BTC, digital assets have a massive opportunity. As such, investors should know which NFTs are most likely to shine in 2021. Rarible is an Ethereum blockchain-based platform for NFT. By designation, the platform helps users create, trade quickly, and collect blockchain secured digital items. This platform is home to thousands of pieces of art, around 20 thousand creators and collectors. The Rarible governance token is $Rari. Rarible The platform is community-driven and possesses unique features like flexible royalties, Rari rewards, auction functions, multiple minting, unlockable items, and many others. The platform is also highly user-friendly. The users can easily create single and multiple collectibles, set the properties, details, and royalties to earn rewards. The Falcon Project The is an entire ecosystem that runs on two blockchains at the same time and includes three interchangeable chains. Falcon Project The ﬁrst chain is the ERC-20 (FNT), which is convenient and familiar to all of us, both users and developers. The second chain is an NFT (Voucher), for the storage of which you will receive a guaranteed profit of up to 40% per annum in FNT tokens. The third chain is the coin (FNC), which optionally allows performing anonymous transactions. In addition, the Falcon Project provides a solution as Proof of Ownership for using NFT certificates for collectibles. The purpose of this platform's introduction is to step into the rare collectible industry, comics, etc. A unique QR code is embedded in each product, and when scanned, the blockchain stores the product. Since the NFTs represent ownership of these products, a user can claim rare collectibles any day. The silver souvenir coin is the first proof of the Falcon ownership model, with the coin holders receiving an NFT certificate. Async Network Launched in February 2020, is a blockchain-based art movement created from collecting and trading digital arts and other programmable arts. Providing the option of digital art makes it easy for art to evolve and react to owners. Async art In this network, users quickly buy masters and layers. Layers are components making up a master, while a master is an edition art piece. By controlling the layers, collectors control and possibly affect artist works. The platform is quite user-friendly, with everything displayed well on the homepage. According to the Async web page, Async has already garnered over a bid volume of $5 million and artist sales of $1 million. ADOR is another one of the NFT platforms which users should watch out for in 2021 and beyond. It allows users to earn from their content on a real-time basis. Its native token $ADORs helps maintain the perfect run of things in the Adors project. It unlocks the platform's fullest potential, it helps maintain the perfect run of things in the ADOR's project. ADOR Their unique edge focuses on Human Expression & Media Experiences with unique utilities and allows users to earn from their content on a real-time basis. It provides opportunities for its users to earn from other options like staking and farming tokens, monthly platform rewards, free RARE-Exclusive NFTs and no-wait members only DROPs! Among creators set to benefit highly include, artists, influencers, celebrities and performers who will benefit even from old and new content. It's quite user friendly with a button that sends the user directly to connect their web3 wallets. Finally, with a robust roadmap that will include consulting, virtual land generation,  streaming services and educational tools; ADOR is setting the stage to be a top contender in the METAVERSE industry. SuperRare is another top functional NFT marketplace, allowing its users to collect, and in the long-run, trade some unique forms of digital artworks. It acts as a network for digital art. SuperRare The network ensures that all its collectibles are authentically created by the artists in the network and ultimately tokenized, making it easy to access them. The art therein is scarce but secured with the blockchain, thus provable and transparent. The marketplace fees for this platform are charged at 3%, paid by the buyer. ETH is the main currency for payment in the SuperRare network. The process of signing up is easy and will require the users to connect with an Etherum wallet. CryptoPunks from Larvalabs Launched in 2017 as the first NFT project, is a collectibles platform designed by Larvalabs and based on the Ethereum blockchain. Currently, there are about 10 thousand collectible characters dubbed punks. Cryptopunks Punks are some art images of 24x24 pixels, created like punky looking persons, apes, or aliens.  Due to the blockchain base, users can be sure that only 10k punks are on the platform. Getting a punk is also relatively easy; you only need to follow the following steps; Get Metamask and connect the platform to your wallet. Buy some amount of Ether. The Cryptopunks network will provide a plugin to bid, buy or sell punks directly from the ETH interface. Final Word Although for over three years, NFTs have existed, they are still lagging in terms of adoption, especially when you compare them to ERC-20 tokens and assets like stablecoins and altcoins. However, the NFT world is growing vastly and very soon. Among the to watch in 2021 include the six ones above. Platforms like Adors, SuperRare, Rarible, Async, and CryptoPunks mostly focus on art and collectibles and will also increase vastly in adoption. Also, Falcon will thrive in the industry by providing digital NFT certificates and vouchers. Digital certificates will be some form of immutable proof of ownership that most people will use. NFT projects

NFT Projects of The Day: Rarible Vs. Falcon Vs. Async Vs. ADOR Vs. SuperRare Vs. CryptoPunks

NASA’s Post-ISS Plans May Not Add Up: Here's Why

Practical LLMs for Real-World Applications

Authors:
(1) Jianhui Pang, from the University of Macau, and work was done when Jianhui Pang and Fanghua Ye were interning at Tencent AI Lab (nlp2ct.pangjh3@gmail.com);
(2) Fanghua Ye, University College London, and work was done when Jianhui Pang and Fanghua Ye were interning at Tencent AI Lab (fanghua.ye.19@ucl.ac.uk);
(3) Derek F. Wong, University of Macau;
(4) Longyue Wang, Tencent AI Lab, and corresponding author. Table of Links Abstract and 1 Introduction 2 Related Work 3 Anchor-based Large Language Models 3.1 Background 3.2 Anchor-based Self-Attention Networks 3.3 Anchor-based Inference 4 Experiments and 4.1 Our Implementation 4.2 Data and Training Procedure 4.3 Evaluation 5 Results 6 Analysis 7 Conclusion, Limitations, Ethics Statement, and References A More Experimental Results B Data Settings 3.1 Background Transformers. LLMs are primarily realized as decoder-only transformers (Vaswani et al., 2017; Touvron et al., 2023a,b), incorporating an input embedding layer and multiple decoder layers. Each layer contains a self-attention network and a feedforward network with normalization modules. Crucially, causal attention masks are employed, allowing tokens to attend only to preceding ones. Self-Attention Networks. Typically for decoderonly LLMs like Llama2 (Touvron et al., 2023b), self-attention networks (SANs) map queries Q, keys K, and values V into an output, as delineated in the following equations, where M denotes an L × L masking matrix, facilitating the current i-th token to attend to only preceding tokens whilst disregarding subsequent tokens during the training and inference phases. Keys/Values Caches. In the application of LLMs, the keys/values caches increase with lengthy prefix texts and continuously generated tokens during the inference phase, such as in question-answering (Saad-Falcon et al., 2023), text summarization (Basyal and Sanghvi, 2023), and machine translation (Pang et al., 2024). The key and value matrices associated with tokens of prefix inputs are cached to avoid recomputation and expedite subsequent token prediction (Radford et al., 2019). Additionally, the model generates the output token-by-token in the real-time inference process, which requires more cache memory to store the newly generated sequence. Therefore, addressing the challenges arising from the ever-expanding texts is crucial for enhancing the efficiency of LLM inference. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Jianhui Pang, from the University of Macau, and work was done when Jianhui Pang and Fanghua Ye were interning at Tencent AI Lab (nlp2ct.pangjh3@gmail.com); (2) Fanghua Ye, University College London, and work was done when Jianhui Pang and Fanghua Ye were interning at Tencent AI Lab (fanghua.ye.19@ucl.ac.uk); (3) Derek F. Wong, University of Macau; (4) Longyue Wang, Tencent AI Lab, and corresponding author. Authors: Authors: (1) Jianhui Pang, from the University of Macau, and work was done when Jianhui Pang and Fanghua Ye were interning at Tencent AI Lab (nlp2ct.pangjh3@gmail.com); (2) Fanghua Ye, University College London, and work was done when Jianhui Pang and Fanghua Ye were interning at Tencent AI Lab (fanghua.ye.19@ucl.ac.uk); (3) Derek F. Wong, University of Macau; (4) Longyue Wang, Tencent AI Lab, and corresponding author. Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2 Related Work 2 Related Work 3 Anchor-based Large Language Models 3.1 Background 3.1 Background 3.2 Anchor-based Self-Attention Networks 3.2 Anchor-based Self-Attention Networks 3.3 Anchor-based Inference 3.3 Anchor-based Inference 4 Experiments and 4.1 Our Implementation 4 Experiments and 4.1 Our Implementation 4.2 Data and Training Procedure 4.2 Data and Training Procedure 4.3 Evaluation 4.3 Evaluation 5 Results 5 Results 6 Analysis 6 Analysis 7 Conclusion, Limitations, Ethics Statement, and References 7 Conclusion, Limitations, Ethics Statement, and References A More Experimental Results A More Experimental Results B Data Settings B Data Settings 3.1 Background Transformers . LLMs are primarily realized as decoder-only transformers (Vaswani et al., 2017; Touvron et al., 2023a,b), incorporating an input embedding layer and multiple decoder layers. Each layer contains a self-attention network and a feedforward network with normalization modules. Crucially, causal attention masks are employed, allowing tokens to attend only to preceding ones. Transformers Self-Attention Networks. Typically for decoderonly LLMs like Llama2 (Touvron et al., 2023b), self-attention networks (SANs) map queries Q , keys K , and values V into an output, as delineated in the following equations, Self-Attention Networks. Q K V where M denotes an L × L masking matrix, facilitating the current i -th token to attend to only preceding tokens whilst disregarding subsequent tokens during the training and inference phases. M L L i Keys/Values Caches . In the application of LLMs, the keys/values caches increase with lengthy prefix texts and continuously generated tokens during the inference phase, such as in question-answering (Saad-Falcon et al., 2023), text summarization (Basyal and Sanghvi, 2023), and machine translation (Pang et al., 2024). The key and value matrices associated with tokens of prefix inputs are cached to avoid recomputation and expedite subsequent token prediction (Radford et al., 2019). Additionally, the model generates the output token-by-token in the real-time inference process, which requires more cache memory to store the newly generated sequence. Therefore, addressing the challenges arising from the ever-expanding texts is crucial for enhancing the efficiency of LLM inference. Keys/Values Caches This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Stories, News & Mentions about FALCONX

Falconx’s Stories on HackerNoon

Falconx's News & Mentions