DeFi yield protocol infiniFi unveils INFI token economic model, with 45% allocated to the ecosystem.

PancakeSwap’s team The Kitchen recently released the "The Flip" proposal on the governance forum, proposing to raise liquidity provider (LP) fee splits for v3 and Infinity across all 11 supported chains to 88%, with the protocol retaining 12% for the Treasury; v2 remains unchanged due to hard-coded fees. The proposal states that based on 2025 trading volume, LPs could earn an additional ~$62 million in fees. Higher splits may attract more liquidity, reducing transaction price impact. Trader fee tiers will stay unchanged; existing Infinity pools will see lower total swap fees due to reduced protocol fees. For CAKE, post-veCAKE emissions are proposed to be cut by ~69% from 22,500 tokens daily to 7,000 daily, with the remaining 7,000 tokens allocated to the ecosystem growth fund Pantry. The Kitchen noted that CAKE’s supply has decreased by 11.5% since Tokenomics 3.0 was passed, with v2 alone sustaining net deflation, and the protocol is on track to meet its 2030 20% supply reduction target ahead of schedule. If approved, implementation will start in November 2026, expected to wrap up by the end of January 2027, rolled out gradually per chain. The final plan will be submitted for a Snapshot vote after incorporating community feedback.

Beating AI Express News reports: Lauren Tan, engineering lead of Grok Bot, is calling for the removal of the vast majority of comments in codebases. She argues that coding agents tend to mimic existing code; a comment explaining a temporary fix could be repeatedly copied by subsequent agents as the correct approach, eventually leading to outdated patches and workarounds spreading across the entire project. Tan previously revealed in an interview that Dune, the development framework used by Grok Bot, has banned code comments. She estimates that 99% of AI-generated comments she has encountered explain irrelevant historical context. For instance, an engineer might only request that a specific approach not be used in a pull request (PR), but the agent will write that line into a comment, leaving it for subsequent agents to treat as a permanent rule. To address this, Tan recommends using the `/no-comments` skill from her open-source toolkit pstack. This tool invokes a dedicated agent to review comments, identify outdated explanations and temporary workarounds, and resolve root issues through refactoring. For example, architectural requirements originally written in comments can be converted into automatically enforced lint rules, or type systems can be used to directly block incorrect implementations.

According to Bybit market data, US-listed optical module and storage stocks were trading lower in pre-market hours, with the following movements: COHR down 0.27%, LITE down 0.66%, AAOI up 0.22%, NOK up 1.52%, MRVL down 0.31%, STX (Seagate) down 1.47%, WDC (Western Digital) down 1.24%, SNDK (SanDisk) down 2.16%, and MU (Micron) down 0.63%.

According to Bybit market data, as Bitcoin rebounded to briefly top $84,000, US-listed crypto-related stocks posted modest pre-market gains, with SBET rising 0.83%, BMNR up 0.64%, GEMI up 0.49%, CRCL up 0.64%, MSTR up 1.49%, COIN up 0.57%, and ASST up 0.94%.

Beating AI Express: Leading edge-side AI R&D firm Liquid AI has launched LFM2.5-24B-A2B, a model focused on enhancing local agent inference and tool calling capabilities. The previous LFM2-24B-A2B generation prioritized operational efficiency; the new version retains a total of 24 billion parameters with only around 2 billion activated per run, while its pre-training data has grown from 17 trillion to 33 trillion tokens, and its context window expanded from 32K to 131K. The new model performs inference before generating responses, and was fine-tuned via reinforcement learning directly on agent frameworks including OpenClaw, Hermes Agent, Pi, and OpenCode, training on continuous tool calls, handling return results, and completing multi-step tasks. In an official demo, a MacBook—no cloud model API required—enabled Hermes Agent to search for trending Hugging Face projects, compile briefings, and act on follow-up queries. Official tests show the new model generates 184 tokens per second on Apple M5 Max, up to 1.7 times faster than comparable MoE models. In agent benchmarks, it scored 48.09 on LiveClaw, higher than Qwen3.6-35B-A3B’s 38.19; its AutomationBench score of 27.69 is over three times that of Qwen. However, it only achieved 51% on the SWE-bench Verified programming test, lower than Qwen’s 66.2%.

According to official announcements, Binance Wallet will open the claim period for a new round of Alpha airdrops at 18:00 Beijing time today. Users holding at least 245 Binance Alpha points are eligible to claim the token airdrop. The airdrop operates on a first-come, first-served basis and will remain open until the airdrop pool is exhausted or the event concludes.