News linked to both this project and an event.
Pump.fun co-founder Alon stated that the initial Callout reward mechanism previously led to low-quality spam. The platform has adjusted its algorithm to reduce the marginal returns for high-frequency posting, emphasizing that rewards are tied to the actual trading volume generated by the content.
Odaily News: Cryptocurrency exchange Gemini has switched its Zcash system operating software from Zebra to Zakura, reducing the time required for nodes to catch up with the network state from nearly a day to just over 6 hours.The switch was made ahead of Zcash's planned NU7 upgrade. The upgrade aims to reduce block times from 75 seconds to 25 seconds, requiring network computers to process transactions at three times the previous frequency.Developers aim to implement the mainnet upgrade on November 5, but Gemini has not yet published performance test results for Zakura under the faster block production frequency. (CoinDesk)
B.AI has announced that the latest Xiaomi MiMo-V2.6 series will begin its exclusive promotional event on September 25 at 15:00 (SGT): MiMo-V2.6-Flash will be available at 90% off, and MiMo-V2.6-Pro will enjoy a 50% off introductory discount. The series supports a 1M ultra-long context window and full multimodal understanding, covering high-frequency Agent and complex reasoning scenarios, enabling developers to implement API calls cost-effectively. Log in to chat.b.ai/chat in advance to obtain your API Key.
Odaily News: On September 23, the B.AI platform announced the official launch of two high-efficiency models from the OpenAI GPT-6 series, GPT-6 Sol and GPT-6 Luna, simultaneously available on both API and Web Chat. The entire series supports a 1.05M ultra-long context window and 128K output, natively supports image understanding and Computer Use, and offers 6 adjustable levels of reasoning intensity.Among them, GPT-6 Sol is designed for complex software engineering and Agent workflows, scoring 68.8% on DeepSWE v1.1 with per-task costs reduced by approximately 80%; GPT-6 Luna is built for high-frequency, large-batch workloads, scoring 66.6% on DeepSWE v1.1, with input pricing as low as $0.10/1M Tokens. Starting today, developers and users can directly access both models through the B.AI platform.
B.AI has announced the full launch of Xiaomi's open-source, native multimodal reasoning large models in the MiMo-V2.6 series, with simultaneous access via both API and Web Chat. The entire series features a 1M token ultra-long context window and comprehensive understanding capabilities across text, images, video, and audio. The MiMo-V2.6-Pro utilizes a sparse MoE architecture with 1.02 trillion total parameters and 42 billion active parameters, specifically engineered for repository-level software engineering, long-horizon agents, and complex multimodal reasoning. The MiMo-V2.6-Flash, on the other hand, is built on an architecture with 309 billion total parameters and 15 billion active parameters, prioritizing high throughput and rapid response for high-frequency office automation and large-scale agent invocation scenarios. Starting today, developers and users can directly invoke both models via the B.AI platform. Visit chat.b.ai/chat to experience them now.
Solana shortened its target slot time from 300 milliseconds to 250 milliseconds on Friday, increasing block production frequency by nearly 17%. The computing resources and data volume available per slot are reduced accordingly, while the network's overall transaction processing capacity remains unchanged.This adjustment is the third of four planned phases in the SIMD-0525 proposal. The window during which validators have block production rights for four consecutive slots has been shortened from 1.2 seconds to 1 second, and the estimated duration of an Epoch, fixed at 432,000 slots, has been reduced from about 36 hours to about 30 hours.The next phase plans to further reduce slot time to 200 milliseconds, but a mainnet launch date has not yet been determined, and the implementation condition is that the block skip rate remains stable. (Decrypt)
B.AI officially announces the arrival of five popular models now available at just 10% of their standard price. Starting September 16 at 17:00 (SGT), Qwen3.8-Flash, Hy3, and MiMo-V2.5 officially join the 10% pricing tier, teaming up with DeepSeek-V4.1-Flash and GLM-5.3-Flash—already available at this rate—to form B.AI’s exclusive five-model discounted bundle. This means developers can seamlessly switch across diverse workloads including inference, coding, multimodal processing, Chinese-language applications, and high-frequency APIs by paying only 10% of the official standard rate, empowering them to build cost-efficient AI infrastructure. Whether for independent development or large-scale team deployments, B.AI makes top-tier compute power accessible, truly delivering a low-cost, highly flexible model calling experience. Start your trial now: https://chat.b.ai/chat
Less than 24 hours remain until the new pricing for the DeepSeek V4 Flash series takes effect (September 10, 12:00 SGT). With upstream price adjustments imminent, B.AI has prepared an exclusive additional perk for global developers to ensure top-tier compute remains accessible. This upcoming benefit aims to further reduce API call costs and unlock greater productivity—whether you're handling high-frequency API calls, building Agent workflows, or executing large-scale production deployments, you can sustain efficient operations at a lower cost. Specific details will be announced shortly. Please stay tuned to B.AI's official updates and don't miss this compute boost. Secure B.AI in advance to build up compute reserves for your Agent development: https://chat.b.ai/chat
Odaily News, Bitget has released its August 2026 transparency report, disclosing the latest progress in building its multi-asset trading ecosystem. In terms of US stock tokens, taking rNVDA as an example, its single-day trading volume reached $38.5 million, with over 4,200 users completing approximately 112,800 trades. Around 36% of this trading volume occurred outside regular US market hours, reflecting that rToken is becoming a core tool for investors to break through traditional trading time constraints and achieve 7×24-hour global asset allocation.On the institutional business front, Bitget has launched an institutional-grade CFD liquidity solution targeting quantitative teams, proprietary trading firms, funds, brokers, and high-net-worth professional traders, supporting high-frequency quantitative trading, spot-futures arbitrage, and automated trading scenarios such as EAs. Meanwhile, the platform has listed FCNs (Fixed Coupon Notes), bringing traditional financial structured products into the tokenized US stock market, making it the first crypto exchange to combine the FCN structure with USDT settlement and US stock rToken delivery. As related products and trading infrastructure continue to improve, Bitget is further strengthening its capabilities in institutional-grade services, multi-asset coverage, and liquidity.Additionally, third-party data shows that Bitget stands out in terms of liquidity and execution efficiency for tokenized stocks and stock perpetual contracts. In DeFiLlama's liquidity benchmark tests covering five stock spot markets — MSTR, SPY, QQQ, CRCL, and NVDA — Bitget recorded the lowest median bid-ask spread of 0.83 basis points and maintained the deepest order book liquidity across all sampled markets. In execution benchmark tests covering 36 stock perpetual contracts and 8 metal and commodity perpetual contracts, Bitget maintained a deep lead on approximately 90% of trading pairs.
Zhipu's official GLM-5.3-Flash will undergo a pricing update shortly. To provide global developers with a more ample preparation window, B.AI has announced a limited-time extension of its free access—until September 12 at 09:59 (SGT), GLM-5.3-Flash will remain completely free on the B.AI platform. Leading in platform usage volume, GLM-5.3-Flash features 320B total parameters and 18B activated parameters, supports a 1M ultra-long context window, and seamlessly combines rapid response with powerful reasoning capabilities, making it a highly cost-effective choice for high-frequency coding, massive data processing, and complex Agent workflows. Additionally, other popular models such as Qwen3.8 Flash, Hy3, and MiMo V2.5 continue to be 100% free on the B.AI platform. B.AI remains committed to supporting global developers with inclusive computing resources, ensuring that frontier model capabilities are truly within everyone's reach.
According to official announcements, United Stables (U) has been listed on Binance Alpha 2.0 today, becoming the platform's first exclusive stablecoin trading pair for Robinhood Chain's trending assets. Users can directly settle assets and execute instant swaps via U within Binance Alpha 2.0 without switching interfaces. This listing bridges Binance CEX's high-frequency trading gateway with the asset potential of Robinhood Chain, providing users with a unified and efficient trading experience.
The GLM-5.3-Flash model has become the most frequently invoked and popular model on the B.AI platform, with cumulative token throughput exceeding 2.41 trillion. As the first native multimodal model in the GLM-5 series, GLM-5.3-Flash features 320B total parameters and 18B active parameters. It employs a hybrid architecture combining sparse and linear attention mechanisms, supports 1M ultra-long context windows, and balances rapid response, powerful reasoning capabilities, and high cost-effectiveness. Starting today, developers can still invoke this model for free via the B.AI platform, covering diverse scenarios such as high-frequency APIs, coding, complex Agents, and ultra-long document processing. Try it now: chat.b.ai/chat
Daily token throughput on the B.AI platform has exceeded 1.1 trillion, marking a 52,133-fold exponential increase compared to the 21.1 million tokens processed on its launch day, April 9. In less than five months, the platform has accomplished an architectural leap in compute infrastructure from the millions to the trillions. Since the free campaign began, daily throughput has grown approximately 242-fold, with a cumulative 14-day total reaching 6.86 trillion tokens, while cumulative newly registered API users have surpassed 196,000. Faced with a continuous influx of massive high-concurrency traffic and complex Agent workflows, B.AI’s underlying infrastructure has proven exceptionally stable. This steep growth curve stands as a definitive vote of confidence from developers globally, who have endorsed B.AI’s compute foundation through real-world code deployment and high-frequency API calls. B.AI will continue to develop as the premier global provider of inclusive compute power and an Agent finance foundation. Log in to B.AI now to seamlessly integrate into a trillion-token compute matrix.
Odaily News: Hyperscale Data, Inc. (NYSE: GPUS), an AI data center and Bitcoin-related enterprise, announced that its wholly-owned subsidiary Gresham Worldwide, Inc.'s radio frequency (RF) business secured approximately $5.4 million in new orders between July and August 2026, primarily from U.S. defense programs and prime defense contractors.These orders cover RF and microwave components supporting radar, electronic warfare, communications, and other mission-critical systems, and are expected to be produced and delivered by the relevant business units of Gresham Worldwide. The company stated that the new orders reflect the continued growth in demand for high-reliability RF technology and critical components in the U.S. defense sector. (PRNewswire)
According to Decrypt, Coinbase announced the official launch of perpetual futures trading in the Base App via Hyperliquid integration. Eligible users can access up to 50x leverage across over 290 perpetual contract markets, including BTC, ETH, tokenized stocks, and commodities. Coinbase Head of Engineering Chintan Turakhia noted that perpetual contracts constitute approximately 75% of current crypto trading volume and are the most requested feature among high-frequency Base App users.
Bitget CFD has officially launched its sub-account feature, aimed at providing institutional clients, professional trading teams, and high-net-worth users with more refined multi-strategy management, achieving 100% physical isolation between different strategies and positions. Sub-accounts share the master account's KYC status with no need for repeated verification. By default, users can create up to 5 independent CFD accounts with one click, supporting the deployment of different strategies—such as EA, high-frequency scalping, long-term trend, and hedging—into separate accounts, ensuring each strategy operates in a clean, independent environment.At the same time, the master account can centrally view real-time equity, margin ratio, current orders, and historical orders of all sub-accounts, while retaining the authority to reset sub-account trading passwords and read-only passwords, catering to professional teams' needs in account management, permission control, and internal risk management. Risk notifications for each sub-account, including margin top-ups and liquidation warnings, will also be independently sent to corresponding emails and app pushes, further enhancing the efficiency of multi-strategy collaboration and risk response.
Odaily News: Cryptocurrency market maker Wintermute plans to invest approximately $1 billion over the next five years in AI infrastructure and high-frequency trading systems, while expanding into equities, commodities, foreign exchange, and prediction markets. The company aims to increase non-crypto revenue to more than 50% of total revenue by the end of 2027. Wintermute founder and CEO Evgeny Gaevoy said the company plans to fund the investment with retained earnings. Wintermute's average daily trading volume this year is around $10 billion, down from approximately $15 billion last year; non-crypto businesses currently account for about 10% of revenue. The investment projects will cover computing power, storage, network, and data center infrastructure, supporting quantitative strategies that rely on large-scale datasets and models requiring continuous training and retraining. Wintermute has already expanded into exchange-traded funds, real-world asset perpetual futures, and prediction markets. Wintermute's U.S. affiliate recently completed its broker-dealer registration, allowing it to trade equities and stock options for its own account and serve as an authorized participant for exchange-traded products. The registration provides a pathway for the company to enter regulated securities markets. (Bitcoin.com News)
According to the official announcement, Bitget CFD has launched Pro mode based on the existing ECN and STP modes, providing independent quote depth for high-frequency and large-volume traders. Currently, this feature is live on the web platform, and the mobile app is expected to open on August 21. Trading instruments in Pro mode uniformly use the .pro suffix; the fee structure and rebate rules are consistent with ECN mode, with a fixed handling fee charged per lot per instrument, and overnight fees charged as normal. This mode is open only to qualified professional clients; users can contact their account managers or official customer service to apply for activation. Previously, Bitget CFD launched liquidity solutions for institutional traders, supporting multi-tier depth aggregation and 100% STP execution. The launch of Pro mode further enriches Bitget CFD's support for professional trading scenarios.
Odaily News - Bitget has officially launched an institutional-grade CFD liquidity solution, targeting quantitative teams, proprietary trading firms, funds, brokers, and high-net-worth professional traders, supporting high-frequency trading, spot-futures arbitrage, and automated trading scenarios such as Expert Advisors (EA). As professional trading institutions continue to demand greater execution efficiency, liquidity, and low latency, this solution aims to provide a more stable and efficient execution environment for large-volume, high-frequency trading.On the execution and liquidity front, Bitget adopts a 100% STP (Straight-Through Processing) model, routing orders directly to external liquidity pools and aggregating multi-level market depth from global Tier-1 banks and non-bank market makers, thereby reducing slippage and market impact during large-order execution. Additionally, trading servers are deployed in core financial data centers such as LD4 in London and TY3 in Tokyo, supporting sub-millisecond order matching via dedicated lines and fiber-optic connections, and offering a FIX API to facilitate institutional clients' integration with existing trading systems, bridging tools, and liquidity aggregation platforms.In terms of fund management, client assets are segregated from platform operating funds, with independent custody accounts, compliance reviews, and third-party audit mechanisms enhancing asset management transparency. The launch of this institutional-grade liquidity solution further strengthens Bitget's CFD backend trading infrastructure, complementing its existing retail-facing products and covering a multi-tiered range of trading needs from retail traders to professional institutions.
According to Yonhap News Agency, Samsung Electronics and Japanese carrier NTT DoCoMo announced that the jointly developed "user-customized communication quality optimization technology" based on AI wireless network (AI-RAN) has been successfully verified. The technology learns user movement paths and service usage patterns through AI, predicts the risk of network quality degradation in advance, and automatically switches to the optimal frequency band and network configuration, achieving a stable experience in scenarios such as seamless HD video streaming. In simulation verification on NTT DoCoMo's commercial network and 5G test network, the frequency of service transmission speed degradation decreased from 13.1% to 7.2%, a reduction of nearly half.