Actualités financières en direct

The prediction market operator is expanding beyond event contracts with leveraged long and short exposure to an index tracking 500 large US companies. Kalshi has launched a perpetual futures contract tied to the US …

The prediction market operator is expanding beyond event contracts with leveraged long and short exposure to an index tracking 500 large US companies. Kalshi has launched a perpetual futures contract tied to the US stock market, expanding its product lineup as the prediction market operator pushes further into traditional financial markets. Today we launched the first stock index perpetual future in America. You can now trade a perp on the US 500, an index of America’s largest companies weighted by market cap. Stock index perps let you trade the broader market with greater capital efficiency and make it easier to… pic.twitter.com/8I1u91EQrK — Kalshi (@Kalshi) October 6, 2026 The contract tracks Kalshi’s US 500 index, which the company said covers 500 of the largest US companies. Traders can take leveraged long or short positions without the contract having a fixed expiration date. Advertisement Unlike conventional futures, perpetual futures do not expire and use periodic funding payments between long and short traders to help keep the contract’s price aligned with the underlying index. Kalshi filed with the Commodity Futures Trading Commission in August to offer the product. The crypto and markets news you actually need—every morning. Daily. Free. Join 34,000+ readers across crypto, finance, and policy. Email address Subscribe free We respect your privacy. Unsubscribe anytime. “Stock market exposure is the next step towards Kalshi becoming a full-service financial exchange, and perps are the best way for our traders to get this exposure,” CEO Tarek Mansour said in a statement. The launch marks another step in Kalshi’s expansion beyond the event contracts that built its business, including markets tied to elections, sports and other real-world outcomes. The company has also been moving into more conventional derivatives markets. Reuters reported last month that Kalshi is preparing to file with US regulators for a perpetual contract tracking West Texas Intermediate crude oil. The expansion puts Kalshi more directly into competition with traditional exchange operators as it builds products spanning prediction markets and financial derivatives.

The Reserve Bank doesn't want consumer prices to come down, but to rise more slowly. It's a common misunderstanding that the ABS is partly responsible for. What is the Reserve Bank trying to achieve …

The Reserve Bank doesn't want consumer prices to come down, but to rise more slowly. It's a common misunderstanding that the ABS is partly responsible for. What is the Reserve Bank trying to achieve by raising interest rates? In recent days, a video has done the rounds on social media of someone asking for help to understand what the RBA is up to. This…

Firelight, a DeFi cover protocol incubated by Sentora, has launched insurance coverage backed by 50.2 million XRP staked on the Flare network. This development marks the introduction of a new utility for XRP within …

Firelight, a DeFi cover protocol incubated by Sentora, has launched insurance coverage backed by 50.2 million XRP staked on the Flare network. This development marks the introduction of a new utility for XRP within the DeFi ecosystem, as staked XRP is used to back coverage for risks such as smart-contract exploits and oracle failures. The total aggregate cap for staked XRP positions is set at $115 million. By leveraging XRP bridged as FXRP, Firelight adds an on-chain protection market to Flare, providing coverage for vaults like Sentora’s USD Protected Vault. Key Takeaways Firelight’s use of staked XRP for DeFi cover suggests increased utility and adoption of XRP. The aggregate cap for staked XRP positions is $115 million, indicating a significant commitment to backing coverage. This development appears to align with market sentiment supportive of a positive outcome for XRP’s long-term prospects. What to Watch Market participants will likely focus on how the introduction of DeFi insurance impacts XRP’s adoption and utility. Additionally, the reaction of key actors such as Ripple CEO Brad Garlinghouse and financial institutions like BlackRock and Fidelity could provide further indications. Observers should also monitor regulatory developments, including any actions from the U.S. SEC, which may influence market sentiment and the pricing of XRP’s all-time high potential by December 2026.

South Asia’s growth has remained more resilient than expected despite the energy shock. How durable is this momentum? It is a broad-based upgrade. We had expected a slowdown because of the energy market disruptions …

South Asia’s growth has remained more resilient than expected despite the energy shock. How durable is this momentum? It is a broad-based upgrade. We had expected a slowdown because of the energy market disruptions elsewhere in the world. What has happened instead is that growth has held up. We expect growth to stay at 6.8 per cent in 2026-27. It has been much more resilient than we had expected, and that’s the whole region, almost the whole region. It’s not just India. To some extent, the growth momentum is very durable because domestic demand remains very robust and resilient. Across the region, governments have taken measures to buffer the impact of high energy prices. Private credit growth in several countries has also been very robust, holding up investment. The question is whether the underlying assumption behind the policy action is right: Is the energy price increase temporary or permanent? That is our downside risk. What could happen if energy prices remain elevated? If refined petrol and refined energy product prices stay high, then within the next few months governments need to begin to unwind some of the policy measures that were meant to buffer a temporary shock. In the long run, they need to be unwound, and energy demand will have to be depressed. Are there any upside risks to the growth outlook? I think the upside risk is in the long term. In the short term, it is probably mostly downside risks. This resilience has been surprising, but the longer resilience gets tested by shocks, whether that’s El Niño, energy prices or some financial market disruption in the rest of the world, the bigger the downside risk. But in the long term, that is where the upside risk is: artificial intelligence (AI) could really unleash productivity gains. What would governments in South Asia need to do to capture these AI-related gains? The most important thing is to create the right enabling environment through sufficiently fast broadband, large energy capacity, reliable internet, a workforce that is skilled not just technically in AI but in the foundational skills that allow people to work with AI, and a good business climate. Firms need to be able to experiment with AI. There needs to be trial and error. Some firms need to expand, others need to contract, and there needs to be reallocation of workers and money. The government’s role is to regulate, and regulation must be clear, predictable, and reliable to take uncertainty out of the system. Uncertainty really damages investment. Can AI become a significant source of employment and wage growth in South Asia? Previous general-purpose technologies have given the economy a boost in all dimensions, both in productivity and employment growth. The internet revolution took a long time to really permeate firms, but now everyone’s using it. Employment has risen dramatically since then. One of the biggest drivers of wages is usually productivity. If you have productivity growth, usually on average, over time, you get wage growth as well. There is also an interesting pivot already happening in labour markets. We had previously established that the number of jobs in the AI expert industry has gone down. But firms are increasingly hiring or posting job ads for workers with AI-augmented skills. So there is a pivot happening in labour markets. How long could it take before the productivity and employment gains from AI become visible in South Asia? Could those gains materialise faster in India, given its relatively rapid AI adoption? Hard to tell. These things take a long time. If you just look at historic evidence, it took decades. Now this may be much faster than previous instances, but it may still take a decade. India is adopting at twice the rate of others. Individuals are adopting it much faster than the rest of South Asia, and firms are adopting it much faster than other Asian countries. Are there any other major downside risks for the growth forecast for South Asia? El Niño is a risk. We looked at past severe El Niños, and it is crop production that significantly falls. Not necessarily gross domestic product — that really depends on the region and the country — but crop production in every country significantly falls. Past El Niños also brought big flooding that doesn’t only affect agriculture; it affects cities as well. There is also the risk of a global correction in AI-related activity. The whole world is currently benefiting from the surge in global AI-related activity. All of East Asia is part of the supply chain, but South Asia is not.

The five-year GPU infrastructure agreement with the AI company Cohere marks a significant revenue milestone for the Nvidia Preferred Cloud Partner. Boost Run, Inc. has locked in a five-year cloud services agreement with AI …

The five-year GPU infrastructure agreement with the AI company Cohere marks a significant revenue milestone for the Nvidia Preferred Cloud Partner. Boost Run, Inc. has locked in a five-year cloud services agreement with AI company Cohere Inc. worth $525.6 million in total contract value, the company announced on October 6, 2026. The deal pushes Boost Run’s cumulative contracted revenue across all agreements past $2.6 billion. Markets noticed. Shares of Boost Run (Nasdaq: BRUN) rose approximately 10 to 11 percent in pre-market trading following the announcement. Advertisement What the deal actually involves Boost Run will deliver dedicated GPU cloud computing infrastructure to Cohere using NVIDIA GB300 NVL72 systems. Initial infrastructure acceptance is expected to begin in early Q2 2027. The agreement also includes conditional terms tied to delivery timelines, with specific performance metrics required to be met by July 15, 2027. Boost Run CEO Andrew Karos framed the deal as a demonstration of the company’s ability to translate its financing into contracted, ongoing revenue streams. The company is recognized as an NVIDIA Preferred Cloud Partner, meaning it builds to NVIDIA Reference Architecture standards. AI, tech, and the markets they move—in one daily briefing. Daily. Free. Join 34,000+ readers across crypto, finance, and policy. Email address Subscribe free We respect your privacy. Unsubscribe anytime. Why $2.6 billion in TCV is worth paying attention to The company also has a prior lease agreement with 10X Infrastructure Partners that expanded its available capacity, giving it the physical GPU footprint to honor multi-year deals at this scale. What to watch from here The most immediate thing to monitor is whether Boost Run meets the July 15, 2027 delivery milestone embedded in the Cohere agreement. CEO Karos referenced a diversified GPU fleet and customer base as a strategic priority.

Treasury tokens keep flowing onto Stellar as the network builds a real-world asset base measured in billions The market cap of tokenized US Treasury bills on Stellar climbed by $96.2 million over the past …

Treasury tokens keep flowing onto Stellar as the network builds a real-world asset base measured in billions The market cap of tokenized US Treasury bills on Stellar climbed by $96.2 million over the past week. The jump adds to a steady run of Treasury inflows on the network. A week of Treasury inflows A tokenized T-bill is a digital token that represents a claim on a short-term US government IOU, or on a fund holding those IOUs. A separate RWA Activity report from the RWA Foundation and Token Terminal, covering the week ending October 5, 2026, logged its own sizable move. Advertisement In that report, tokenized T-bills on Stellar added $81.2 million to their market cap. Total tokenized funds on the network grew by $90.9 million that week, with Treasuries doing most of the heavy lifting. Ethereum’s tokenized T-bills grew by $93.2 million over the same period, edging out Stellar’s weekly gain. The news moving money, markets, and the world—before your day starts. Daily. Free. Join 34,000+ readers across crypto, finance, and policy. Email address Subscribe free We respect your privacy. Unsubscribe anytime. Issuer-level figures in the research reportedly credit J.P. Morgan with $81.4 million in T-bill growth. Franklin Templeton reportedly contributed $72.6 million. From side project to a $3.47 billion base The network’s market cap for these assets grew from an estimated $500 million in early 2025 to more than $3 billion by June 2026. As of early October 2026, distributed real-world assets on Stellar stood at approximately $3.47 billion. Franklin Templeton’s BENJI product has been central to the narrative. It launched on Stellar in April 2021, long before tokenization became a standard line in bank strategy decks. What the inflows mean for the tokenization race Ethereum still holds the overall lead in total Treasury token volumes, according to the research, and it outgrew Stellar in the October 5 reporting week. Weekly market cap figures can swing as funds rebalance, redeem, or shift between chains, so a single strong week does not guarantee a trend line. When a handful of large issuers account for much of the growth, a strategy change at one firm could noticeably dent a network’s numbers.

The French lab's trillion-parameter open-weight model enters limited preview, with public weights scheduled for October 27 Mistral AI has a new flagship. Its name is Mistral Large 4, and its nickname, officially, is “Le …

The French lab's trillion-parameter open-weight model enters limited preview, with public weights scheduled for October 27 Mistral AI has a new flagship. Its name is Mistral Large 4, and its nickname, officially, is “Le Chonk.” The French company unveiled the model on October 6, 2026, pitching it as the strongest open-weight AI system built outside China. Mistral also says it outperforms some Chinese rivals. Anyone can judge that claim once the weights go public on October 27. What Mistral actually built Mistral Large 4, or ML4, carries 1 trillion parameters. Only 49 billion of them are active at any given moment. That gap comes from the model’s design. ML4 is a Mixture-of-Experts system, or MoE. Picture a hospital rather than a single general practitioner. When a patient walks in, they get routed to the right specialists instead of every doctor in the building. An MoE model works the same way, sending each request to a small subset of its internal “experts” rather than firing up the whole network. Advertisement ML4 is also natively multimodal. It was built from the ground up to handle more than plain text, rather than having image capabilities bolted on later. Training took approximately two months. Mistral used about 4,000 NVIDIA Grace Blackwell GPUs housed in its own European data centers. AI, tech, and the markets they move—in one daily briefing. Daily. Free. Join 34,000+ readers across crypto, finance, and policy. Email address Subscribe free We respect your privacy. Unsubscribe anytime. The benchmark claims, read carefully According to Mistral, ML4 scored 38 points on the Artificial Analysis Intelligence Index. The company says that is the highest score for any non-Chinese open-weight model. MiMo-V2.6-Pro, one Chinese open model, scored 46 on the same index. That puts it comfortably ahead of ML4’s 38. Mistral is targeting ML4 at specific workloads. These include cybersecurity, finance, coding, manufacturing and visual grounding, which is the ability to connect language to specific elements within an image. The model supports more than 160 languages. That list covers every official language of the European Union. Pricing and availability For now, ML4 sits in a limited preview. Full access is restricted to select developers and expert users. That changes on October 27, 2026, when Mistral plans to release the model’s weights publicly. For those who prefer the rental route, Mistral has posted API pricing. Input costs $1.36 per million tokens, and output costs $4.18 per million tokens.

Some companies are preparing for an AI-fueled chip squeeze that could push up prices far beyond the data center boom alone, Mary Daly, president of the Federal Reserve Bank of San Francisco, tells Axios. …

Some companies are preparing for an AI-fueled chip squeeze that could push up prices far beyond the data center boom alone, Mary Daly, president of the Federal Reserve Bank of San Francisco, tells Axios. Why it matters: The Fed can usually look through supply… Some companies are preparing for an AI-fueled chip squeeze that could push up prices far beyond the data center boom alone, Mary Daly, president of the Federal Reserve Bank of San Francisco, tells Ax…

The probability of an October Fed rate hike has fallen to 19% from about 51% a week earlier, according to CME FedWatch, after September payrolls rose by only 29,000. Bitcoin’s recent price rally has …

The probability of an October Fed rate hike has fallen to 19% from about 51% a week earlier, according to CME FedWatch, after September payrolls rose by only 29,000. Bitcoin’s recent price rally has been supported by lower expectations for a Federal Reserve rate hike, but the crypto asset needs stronger market flows to confirm a move toward $90,000-$93,000, according to Lacie Zhang, research lead at Bitget Wallet. CME FedWatch showed the probability of an October hike at about 19% on Oct. 6, down from 51% a week earlier following September payroll growth of just 29,000. A decline in Treasury yields and further evidence of softer inflation could give Bitcoin additional room to rise, Zhang noted. Advertisement “ETF demand remains supportive, but it is not yet strong enough on its own to force a breakout,” Zhang told Crypto Briefing. “US spot Bitcoin ETFs attracted about $2.6 billion in September and approximately $134 million across the first two trading sessions of October. Bitcoin’s repeated failure to hold above $87,000 suggests that profit-taking and existing supply are still absorbing institutional demand.” Bitcoin ETF flows became increasingly choppy heading into October after hitting a $999 million daily net inflow peak on Sept. 21. The shift marked a sharp slowdown in institutional buying, with investors alternating between fresh allocations and profit-taking rather than maintaining the strong accumulation seen in late September. The pattern continued into October, with net inflows of $103 million on Oct. 1 and approximately $190 million on Oct. 2 before flows swung to $90 million in net redemptions on Oct. 5, per Farside Investors. The crypto and markets news you actually need—every morning. Daily. Free. Join 34,000+ readers across crypto, finance, and policy. Email address Subscribe free We respect your privacy. Unsubscribe anytime. BlackRock’s IBIT remains the dominant Bitcoin ETF, with cumulative net inflows approaching $66 billion. The fund has consistently brought in new money even on days when the ETF market records net outflows. Zhang said a convincing breakout would require sustained ETF inflows, stronger spot buying and a daily or weekly close above roughly $87,400. The next upside targets are $90,000 and $93,000, while $84,000 and $82,000 are key downside levels, she added. “The main risks are a stronger-than-expected CPI or PPI report, renewed oil-driven inflation, hawkish Fed guidance or another rise in long-term yields. Any of these could revive rate-hike expectations and push Bitcoin back toward $84,000, with $82,000 as the next major support,” she stated. Bitcoin fell back below $86,000 at press time after reaching $86,698 earlier this morning, as most altcoins also declined. The last time BTC traded above $90,000 was in January.

In its infancy, the crypto sector was bent on creating entirely new types of assets. First came Bitcoin and its derivatives, then Ether and its competitors. Governance tokens, NFTs, and memecoins all followed a …

In its infancy, the crypto sector was bent on creating entirely new types of assets. First came Bitcoin and its derivatives, then Ether and its competitors. Governance tokens, NFTs, and memecoins all followed a formula: create a new digital asset class and build a market around it. As the industry has matured, however, its focus has shifted to creating new markets instead of new assets. We have now three major examples in the form of prediction markets, oil and gold perpetuals on Hyperliquid, and pre-IPO perpetuals. These products created real-time, continuously tradable markets around existing things (news, commodities, private firms) that previously lacked them. It turns out that blockchain technology is very well suited for expanding the range of things that we can price. And that might end up being more relevant to the future of finance than the ability to create new, digital assets. Annabelle Huang is the co-founder and chief executive officer of Altius Labs, an infrastructure company that designs high-performance blockchains. From new assets to new markets