BREAKING: Peru plans to use @Starlink to connect 40,054 public schools that currently have no internet. The project will reach remote, high-altitude and border regions, using solar power where needed. The goal is to connect 100% of Peru’s public schools by 2028.

Season 1, Week 21 is in the books. → 604,054 AMPs distributed → 2,240 wallets earned AMPs → 52,906 AMPs earned through referrals Another week of traders earning on Decibel. →

🖼️ 24 HOURS OF ART 📊 September 19, 2026 🎨 Pictured: 24 Hour Highlights - 'Fidenza #717#' by @tylerxhobbs, acquired via @gondixyz - 'Where My Vans Go #119#' by @DrifterShoots, collected by @ivoDYOR - 'Light Years #054#' by @dmitricherniak - 'STEADY' - 'REMNANT #132#' - 'EXIT STRATEGY' (@punk6529 Memes) by @XCOPYART - 'Disaster Girl' by @esraeslen, acquired via @GalerieMet - 'the rise and fall' by @toadswiback x @Xer0x_XYZ x @luispon_c - 'LORE #26#' by @hahajohnx, collected by @durtis ✍️ NEW: Weekly Dose of Art! 'Transcendent Slaughter: Die With The Most Likes' Poison Candy' - deep dive interview of @toadswiback by @justinwetch - read it via the Articles tab

Intel's 32GB Arc Pro B70 finally got a proper Qwen3.8-27B Local AI workout. So this is why the 32GB VRAM is important, even from an Intel GPU. Here is Gigazine's setup. Windows 11 using Unsloth Desktop + llama.cpp ... 🧠 Qwen3.8-27B Q4_K_XL 💾 30.0GB VRAM + ~1GB shared RAM 🚀 23.4 tps Then they tried Q8: 🧠 Qwen3.8-27B Q8_K_XL 💾 30.2GB VRAM + ~1.5GB shared RAM ⚡ 15.9 tps That's a full 27B-class model running at very usable speeds on an Intel GPU. And also tested text-to-image models. 🎨 Z-Image-Turbo 1024×1024 / 8 steps ➡️ ~5.3 sec/image after warmup 🎥 MiniMax H3 video also ran …but these are much slower, demonstrating where Nvidia's more mature AI software stack still matters. The hardware setup... 🎮 Arc Pro B70 💾 32GB GDDR6 ⚡ 608 GB/s bandwidth 🔌 230W And here's the interesting value angle. In Japan, the tested ASRock B70 was selling for: 💰 ¥298,054 while many RTX 5090s were selling above: 💰 ¥900,000 ⚠️ That's Japan-specific street pricing, NOT a universal B70-vs-5090 price comparison. But THIS is why the B70 interests me for Local AI. It won't beat a 5090 but it's giving you 32GB of VRAM at a much lower entry point. 23.4 tok/s on Qwen3.8-27B Q4 good enough for you? 🔗 GIGAZINE review in ALT

the new launchpad design is live: it's the only place where you can launch a token that pays holders in ethereum:0x07f5b6823751c2e2cd4560f28af75ff887102241. like if it was paired with ethereum:0x07f5b6823751c2e2cd4560f28af75ff887102241. every token created here distributes ethereum:0x07f5b6823751c2e2cd4560f28af75ff887102241 and burns $PONSION. $GRANDKID was just the first test token and it has already burned 0.324% of the $PONSION supply. by itself. five tokens in. 3,706,054 $PONSION gone forever. $PONSION: 0x7a7ed21898ba3c211a3387f598955a2cb8ed3bad $GRANDKID: 0x5a57aa711bd74ff03c228c1fa01f15c37f995944

Weekly Major Events on Calendar! 📅 Aug 24 🔵 X-Agent(@XAgent_official) Launches 2026 AI MCP Hackathon. 🔵 $SOSO unlocks 25.96M tokens (worth $8.29M, 7.592% of the circulation). 📅 Aug 25 🔵 @TermMaxFi ($TMX) TGE. 🔵 @HeebooOfficial ($HEEBOO) public sales starts. 🔵 $COAI unlocks 8.32M tokens (worth $2.74M, 4.424% of the circulation). 📅 Aug 26 🔵 $HUMA unlocks 458.76M tokens (worth $10.29M, 26.467% of the circulation). 🔵 $ALT unlocks 246.59M tokens (worth $1.68M, 3.581% of the circulation). 🔵 @umia_finance ($UMIA) token sale starts. 📅 Aug 27 🔵 Bitcoin Asia 2026 in HongKong on August 27-28. 📅 Aug 28 🔵 $XPL unlocks 297.22M tokens (worth $30.97M, 11.054% of the circulation). 🔵 $BIRB unlocks 27.08M tokens (worth $1.56M, 9.503% of the circulation). 📅 Aug 29 🔵 $HYPE unlocks 9.92M tokens (worth $745.24M, 4.458% of the circulation).