News linked to both this project and an event.
Odaily News - AMD has announced "Day 0" support for Qwen3.8 27B, the latest-generation model from Alibaba's Tongyi Qianwen (Qwen) series, enabling developers to run this large-scale open-source AI model locally on AMD hardware on the very day of its release.AMD stated that Qwen3.8 27B is a 27B-parameter intensive model suited for local AI development, continuing the Qwen series' optimization focus on code generation, practical work tasks, scientific research, and long-context AI applications. The model can run via the open-source inference framework llama.cpp on AI PCs and workstations powered by AMD processors, or on a single AMD 32GB graphics card, while also supporting AMD hardware platforms with over 24GB of variable graphics memory (VGM) or VRAM capacity.AMD's preliminary tests show that Qwen3.8 27B delivers strong local inference performance on AMD platforms: up to 24.5 tokens/second on the AMD Ryzen AI Max+ 395 processor, and up to 51.8 tokens/second on a single Radeon AI PRO R9700 GPU. The tests were conducted on Windows systems using the llama.cpp Vulkan backend with multi-token prediction (MTP) optimization enabled. AMD noted that actual performance still has room for improvement as further software and model optimizations are rolled out.
It was previously reported that Grok 4.6 went live. Latest benchmark information shows Grok 4.6 ranks first in three benchmarks: GDPVal-AA v2, AA-Briefcase, and Harvey LAB, and second in four benchmarks: CursorBench, FrontierCode, APEX-Agents, and APEX-SWE. The specific scores are GDPVal-AA v2 1753, AA-Briefcase 1577, Harvey LAB 15.8%, CursorBench 69.9%, FrontierCode 61.3%, APEX-Agents 57.5%, and APEX-SWE 56.4% respectively. It surpasses Grok 4.5 High in every listed benchmark and ties with GPT-5.6 Sol Max at 61 points on the Artificial Analysis Intelligence Index. Grok 4.6 is available in Cursor and Grok Build starting today, with usage doubling in the first week.
: Bitcoin News posted on the X platform stating that Foundation has released KeyOS 1.3.1 for Passport Prime, adding a mnemonic generation feature. Users can input the first 11 or 23 words of a BIP39 mnemonic, and the device will generate a valid final word that meets checksum requirements. Users can manually generate mnemonics by rolling dice, flipping coins, or other methods, and complete mnemonic generation directly on the Passport Prime without the need for external tools. Foundation stated that AI security reviews conducted using Claude Fable Max and ChatGPT 5.6 Sol Extra High found no serious security issues.
B.AI welcomes multiple major updates this week. In terms of benefits, new users can log in with one click using Bitget Wallet, Binance Wallet, or imToken wallet to immediately claim 1 million free Credits; logging in via an invitation code grants an additional 300,000, stacking with the exclusive wallet login bonus for a cumulative maximum of 1.3 million, while inviters simultaneously enjoy permanent rebates on friends' recharges and subscriptions. Recharge discounts are released simultaneously, with BNB Chain channels enjoying an exclusive 1:1 equivalent quota bonus, while other multiple payment methods enjoy a 1:0.5 bonus; a single user can receive up to $100 equivalent points. On the model side, new breakthroughs continue as the "Choose Your Own Provider" lineup expands significantly this week, adding the Nebula channel and launching historic call discounts as low as 90% off; meanwhile, Alibaba's flagship Qwen model Qwen3.8-Max officially lands on the platform and is now open for limited-time free experience. Currently, the choose-your-own-provider mode fully covers global mainstream large model series such as Claude, GPT, and Gemini; combined with recharge bonuses and multiple discount tiers, it continues to deliver extreme cost-performance computing power for global developers and enterprise users. Log in now at chat.b.ai/chat to start your journey of intelligent creation.
According to Reuters, citing people familiar with the matter, Alibaba plans to introduce commercial terms next week for the next-generation open-source, open-weights AI model Qwen3.8-Max, proposing to require revenue sharing from customers who sell the model as a service and reach a certain scale. The report states that Alibaba previously mainly charged for model usage hosted on its cloud platform, while most open-source models can be used for free when deployed in customers' own data centers.
The Qwen App and PC client have launched new features, introducing multiple new functions such as Thinking & Research, Scheduled Tasks, Office Assistant, Agent Marketplace, and Voice Calls, while supporting Alibaba's latest flagship model Qwen3.8-MAX. All users can use the above features for free.
Alibaba Cloud announced that the Qwen 3.8-Max model received integration support from mainstream AI developer platforms including OpenRouter, OpenCode, Vercel, and DeepInfra on its first day of release. This rapid ecosystem response indicates high interest from the developer community in the new model, with application development underway based on the Qwen 3.8 series. Officially, trial access to Model Studio and Qwen Cloud was simultaneously opened to developers, further lowering the barrier to entry and accelerating the adoption of the new model within the ecosystem.
Alibaba Qwen's latest generation 2.4 trillion-parameter flagship model Qwen3.8-Max has officially launched on the B.AI platform. The model achieves comprehensive improvements in coding, real-world office scenarios, long-horizon tasks, and multimodal agents, and is now live on B.AI with a limited-time free trial. To celebrate the launch, B.AI offers multiple stacked benefits: New users logging in with Binance Wallet, Bitget Wallet, or imToken will immediately receive 1 million free Credits; Top-ups enjoy generous bonuses, with BNB Chain top-ups receiving a 1:1 bonus and other methods 1:0.5, up to a maximum extra $100 per user; Invite friends to register to get 300,000 Credits for them, stacking with the login bonus for a cumulative maximum of 1.3 million, while inviters simultaneously enjoy rebates from friends' top-ups and subscriptions. Limited-time benefits, first come, first served. Log in now at https://chat.b.ai/chat to experience the charm of the new-generation flagship.
According to official news, Qwen releases Qwen3.8-Max, 2.4T parameters, weights to be open-sourced next week Alibaba's Qwen officially releases the most powerful model Qwen3.8-Max, with a parameter scale reaching 2.4T. The model's weights will be open-sourced next week. This is the first time Qwen has open-sourced a Max-level model. Also being open-sourced simultaneously is Qwen3.8-27B.
Odaily News: Alibaba has released its latest flagship model, Qwen3.8-Max, which features 2.4 trillion parameters. The company claims that it performs close to leading global AI models such as Anthropic on certain benchmark tests, surpassing the recently released Chinese model Kimi K3.Alibaba stated that Qwen3.8-Max has improved capabilities in automated programming and long-horizon task execution, and in internal tests, it can independently complete software engineering projects lasting up to 16 days. The model adopts parameter activation optimization technology, invoking only a portion of its parameters during operation to reduce computational costs and latency.Alibaba plans to open-source the Qwen3.8-Max weights next week, alongside the Qwen3.8-27B weights. (Bloomberg)
According to options expiry data released by [email protected] (@BTC__options), a total of 19,000 BTC options expired this week, with a Put Call Ratio of 0.89, Max Pain at $64,500, and a notional value of $1.2 billion; 125,000 ETH options expired, with a Put Call Ratio of 1.25, Max Pain at $1,875, and a notional value of $230 million. In terms of market conditions, BTC briefly broke through 66K this week but failed to hold. The 65K to 80K range is a high-volume trading zone from the early year's rally, indicating significant upside pressure. Overall IV has retreated to the 35% level, leaving fewer trading opportunities. Notably, the ETH put option ratio has exceeded 1 for 6 consecutive weeks, the longest duration on record, indicating sustained high bearish sentiment in the market. However, some traders are also taking the opportunity to sell Puts to buy the dip. Cryptocurrencies have experienced an 8-month bear market. From a cyclical perspective, the second half of the year may see a certain rebound.
据 Forbes 报道，微软针对 Windows 11 版本 25H2 及 24H2 发布了紧急带外安全更新（KB5121767），以修复此前 7 月补丁星期二更新（KB5101650）在部分戴尔设备上引发的性能问题及系统关机故障。 该问题源于最新 Windows USB-C 连接管理器接口与英特尔 Innovation Platform Framework 处理器驱动程序之间的冲突，已知受影响设备包括 Dell Pro Max 14/16 Premium 及 Dell XPS 17 9720/9730 等型号。微软表示，此次更新仅建议受影响的戴尔用户安装，未受影响的设备无需任何操作。
Claude announced that effective July 20, Fable 5 will be officially included in the Max and Team Premium subscription plans, and users will receive usage rights equivalent to 50% of the full quota. Pro and Team Standard users can still access Fable 5 by using credits and will receive a one-time $100 quota compensation. Claude stated that due to the difficulty in accurately predicting the scale of demand after Fable 5's launch, a phased rollout strategy was previously adopted, and the scope of subscriber access was expanded multiple times as compute resources increased. This process brought uncertainty to some users, so this adjustment aims to allow users to more clearly understand the model usage rights included in the subscription plans. In the future, continued investment will be made to expand compute capacity, and Fable 5's access policy will be continuously updated based on capacity conditions.
Odaily AI infrastructure project Venice has announced an update to its VVV and DIEM tokenomics model, introducing a new programmatic buyback and burn mechanism and gradually increasing the DIEM supply target from 38,000 to 40,000 tokens.Venice stated that VVV is its core ecosystem asset, an ERC-20 token issued on the Base chain. After users purchase and stake VVV, they can earn rewards, unlock Venice Pro benefits, and use it to mint DIEM. Previously, a portion of platform revenue was used for market buybacks and the burning of VVV.This update first expands the sources for VVV burning. Previously, Venice had already implemented an automated VVV buyback and burn mechanism through Pro, Pro+, and Max subscription revenues. Now, this mechanism will cover API credit purchases: for every $100 worth of Venice API credits purchased by a user, $5 will be used to buy and burn VVV. This process will be executed automatically. As API usage grows, more VVV will be removed from circulation.Venice stated that the new API burn mechanism will be tracked separately on the official burn page, displayed distinctly from the burn records generated by subscription revenue.Concurrently, Venice will adjust the DIEM supply target. DIEM is the second-generation token in the Venice ecosystem, used to provide access to AI model credits. Each DIEM corresponds to $1 worth of daily Venice AI credits.DIEM can only be minted by staking and locking VVV. Users can use or sell these AI credits. Previously, the DIEM supply target was long fixed at 38,000 tokens.Starting August 3rd, this target will be gradually raised to 40,000 tokens, adding up to 2,000 new DIEM minting capacity to the market. As each newly minted DIEM requires locking more staked VVV, this will correspondingly increase the daily supply of AI API credits.Venice indicated that as the DIEM supply approaches the target, the minting cost (Mint Rate) will gradually increase. This increase in the supply target will provide users with greater minting capacity while maintaining the ecosystem's supply-demand mechanism.This adjustment shows that Venice is strengthening its token economic flywheel through a path of "AI service demand growth → increased API revenue → continuous VVV burning → ecosystem value capture."
Odaily Odaily News According to MAX monitoring, on July 16, the Cascade CLS treasury suspectedly experienced a security vulnerability, resulting in approximately $1.3 million in user fund losses. The platform has suspended all trading and withdrawals and has invited SEAL 911 and other third-party security teams to investigate and handle the incident. Cascade is a 24/7 multi-asset perpetual contract platform headquartered in New York, targeting the US market. It supports deposits via Arbitrum USDC or bank accounts and is currently still in an invitation-only private testing phase.
According to the official announcement, HTX has launched a “Four-Tier Bonus Campaign” for new users. From now until June 12, 2024, at 19:59 (UTC+8), new platform users who complete registration, deposit funds, trading, and referrals will all be eligible for rewards. The campaign includes the following four components: 1. During the campaign period, newly registered users who sign up and complete designated tasks will receive chances to participate in a lucky draw, with prizes including an iPhone 17 Pro Max and XAUT tokens. 2. During the campaign period, new users who achieve a daily spot trading volume of ≥100 USDT or a derivatives trading volume of ≥500 USDT will earn one “check-in.” Users can unlock mystery-box airdrops based on their total number of check-ins. 3. New users will be ranked according to their spot trading volume, competing for a shared prize pool of 30,000 USDT—higher rankings yield larger rewards. 4. During the campaign period, both new and existing platform users who invite friends to register—and whose invited friends complete designated tasks—will receive a 20 USDT reward per successful referral, with a maximum individual reward cap of 300 USDT.
At the Alibaba Cloud Summit, the Qwen3.7-MaX model was launched, enhancing core model capabilities—including language understanding and generation, logical reasoning and computation, knowledge retention and commonsense reasoning, and instruction following and alignment—and supporting multiple Harness frameworks. (Jin10)
Anthropic's Claude has announced a partnership agreement with SpaceX to significantly increase its computing capacity and raise usage limits for Claude Code and Claude API. This includes doubling the 5-hour rate limit for Claude Code in Pro, Max, and Team subscription plans; removing Claude Code rate limits during peak hours for Pro and Max plans; and substantially increasing the Opus model API rate limit. Claude will also utilize the full computing resources of the SpaceX Colossus 1 data center, with plans to add over 300 megawatts of deployment capacity within a month.
According to an official announcement, BAIclaw officially launched its “Sun Brother Brain” feature on May 1. Users can now one-click load the “Justin Sun Perspective” Skill to experience firsthand the deep integration of cutting-edge large language models with top-tier Web3 trading logic. To support deeper on-chain interactions, the platform has simultaneously introduced two premium subscription plans—Plan Pro and Plan Max—offering enhanced execution bandwidth and a smoother trading experience. Additionally, a limited-time launch promotion is now live: new users who make their first deposit of at least $5 will receive a bonus of 5,000,000 points. This update marks B.AI’s further advancement in integrating AI with the on-chain ecosystem—invite you to enter B.AI now, sync with elite intelligence, and seize every wealth opportunity in the Web3 market.
Odaily Odaily News According to official sources, Gate Alpha officially launched its May Day special event at 14:00 (UTC+8) on April 30. Users can trade designated hot tokens to participate in sharing a total airdrop reward worth over $100,000. This event features a dual reward mechanism for basic and advanced tiers: For the basic tier, users can earn one chance to draw a regular mystery box for every 200 USDT traded, with a total prize pool of approximately $30,000, a 100% win rate, and prizes including an iPhone 17, GT tokens, and Alpha fee rebate vouchers. For the advanced tier, once users have traded up to 2,000 USDT, every additional 500 USDT traded grants one chance to draw a premium mystery box, with a total prize pool of approximately $76,000, and prizes including an iPhone 17 Pro Max, substantial GT tokens, and substantial Alpha fee rebate vouchers. Additionally, inviting friends to complete their first trade of 100 USDT will grant extra rewards to both the inviter and the new user. All rewards are distributed in real-time and are available only while supplies last.Gate Alpha now supports popular public chains including SOL, ETH, Gate Layer, BNB Chain, Base, SUI, ARB, World Chain, AVAX, Polygon, LINEA, ZK, OP, and Berachain. It also enables seamless cross-chain token trading through a contract address search function, unlocking cross-chain trading pathways and offering users all on-chain tokens at their fingertips.