OpenSourcePulse

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9208 📅 Released: 2026-05-18 08:22 UTC 📊 Assets: 28 files 📝 What's New: <details open> sycl: route small f32 matmuls to oneMKL, bypass oneDNN (#22150) Signed-off-by: Chun Tao <chun.tao@intel.com> Co-authored-by: Chun Tao <chun.tao@intel.com> </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9203 📅 Released: 2026-05-18 00:18 UTC 📊 Assets: 28 files 📝 What's New: <details open> cmake : fix LLAMABUILDUI logic (#23190) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS Intel (x64) 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9202 📅 Released: 2026-05-17 23:13 UTC 📊 Assets: 28 files 📝 What's New: <details open> cmake : do not install conversion script (#23204) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS Intel (... 🔗 [View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 MasterHttpRelayVPN-RUST 🏷 Category: Community Suggested 🔖 Version: v1.9.29 📅 Released: 2026-05-17 17:11 UTC 📊 Assets: 16 files 📝 What's New: <div dir="rtl"> • رفع خطای JSON در Code.gs نسخه 1.9.28 (PR #1265، #1245، #1253، #1261). مسیر ساده Code.gs ... 🔗 View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9196 📅 Released: 2026-05-17 16:49 UTC 📊 Assets: 28 files 📝 What's New: <details open> vulkan: Support unaligned tensors for ROPE (#22637) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS Intel... 🔗 [View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9193 📅 Released: 2026-05-17 14:10 UTC 📊 Assets: 28 files 📝 What's New: <details open> server : honor --embd-normalize CLI arg (#23125) The --embd-normalize flag was registered only for the embedding and debug examples, so llama-server rejected it and the /embedding handler used a hard-coded default of 2 (L2). Add LLAMAEXAMPLESERVER to the flag's example set and read params.embdnormalize as the handler's default. The per-request "embdnormalize" body field con... 🔗 View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9190 📅 Released: 2026-05-16 23:44 UTC 📊 Assets: 28 files 📝 What's New: <details open> server: (router) alloc tmp buffer on heap (#23159) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS Intel ... 🔗 [View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9189 📅 Released: 2026-05-16 21:20 UTC 📊 Assets: 28 files 📝 What's New: <details open> server: skip device enumeration in router mode to avoid creating CUDA primary context (#23137) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9186 📅 Released: 2026-05-16 19:22 UTC 📊 Assets: 28 files 📝 What's New: <details open> sync : ggml </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS Intel (x64) 📥 Download Release

🚀 New Release Alert! 📦 MahsaNG 🏷 Category: Community Suggested 🔖 Version: v16-(1405-2-25) 📅 Released: 2026-05-16 17:20 UTC 📊 Assets: 5 files 📝 What's New: add MasterDNS add GooseRelay add FlowDriver add Psiphon CDN-Fronting add Psiphon LAN sharing add "type manually HTTP" "psiphon-alone" now run without v2ray config DNS scanner now shuffle ip fix crashes related to getting config & psiphon service fix socks bug when import from link menu UI improved 🔗 View on GitHub 📥 Download Release)

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9181 📅 Released: 2026-05-16 17:06 UTC 📊 Assets: 28 files 📝 What's New: <details open> vendor : update cpp-httplib to 0.45.0 (#23103) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS Intel (x64... 🔗 [View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 MasterHttpRelayVPN-RUST 🏷 Category: Community Suggested 🔖 Version: v1.9.28 📅 Released: 2026-05-16 15:11 UTC 📊 Assets: 16 files 📝 What's New: <div dir="rtl"> • پایپ‌لاین شدن pollهای Full mode و سریع‌تر شدن تونل (PR #1115 از @yyoyoian-pixel). کلاینت Full-mode حالا چند poll هم‌زمان و adaptive نگه می‌دارد تا وقتی یک پاسخ Apps Script یا tunnel-node دیر می‌رسد، بقیه مسیر معطل نماند. برای جلوگیری از به‌هم‌ریختن uploadها، هر write شمارهٔ ترتیبی (wseq) می‌گیرد و tunn... 🔗 View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 MasterHttpRelayVPN-RUST 🏷 Category: Community Suggested 🔖 Version: v1.9.27 📅 Released: 2026-05-16 13:55 UTC 📊 Assets: 16 files 📝 What's New: <div dir="rtl"> • رفع deploy نشدن exit_node.ts روی Deno جدید (#1197، #1120). نسخه‌های جدید deno serve / Deno Deploy انتظار دارند entrypoint به شکل export default { fetch: ... } باشد. exit_node.ts قبلاً یک تابع async را مستقیم default-export م... 🔗 View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 MITM-DomainFronting 🏷 Category: Community Suggested 🔖 Version: MITM-DomainFronting v19 📅 Released: 2026-05-16 11:46 UTC 📊 Assets: 0 files 📝 What's New: Minor tweaks 🔗 View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 MITM-DomainFronting 🏷 Category: Community Suggested 🔖 Version: MITM-DomainFronting v18 📅 Released: 2026-05-16 00:30 UTC 📊 Assets: 0 files 📝 What's New: fastly and github are blocked on most ISPs, but you can clone the entire project (with github desktop) instead of downloading a single file (push/pull also work without any problems) 🔗 View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9173 📅 Released: 2026-05-15 23:47 UTC 📊 Assets: 28 files 📝 What's New: <details open> ci : fix release symlinks (#23119) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS Intel (x64) 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9172 📅 Released: 2026-05-15 22:40 UTC 📊 Assets: 28 files 📝 What's New: <details open> webui: Use lowercase hash for HF checksum check (#23107) </details> macOS/iOS: - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) - macOS ... 🔗 [View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 llama.cpp 🏷 Category: AI 🔖 Version: b9169 📅 Released: 2026-05-15 21:29 UTC 📊 Assets: 28 files 📝 What's New: <details open> mtmd: add chunks and fix preproc for qwen3a (#23073) mtmd: add chunks and fix preproc for qwen3a add attnmask * limit mtmdchunk size (avoid blow up memory) correct audio tokens re-order the setinput case * remove attnmask </details> macOS/iOS: - macOS Apple Silicon (arm64) 📥 Download Release

🚀 New Release Alert! 📦 MasterHttpRelayVPN-RUST 🏷 Category: Community Suggested 🔖 Version: v1.9.26 📅 Released: 2026-05-15 14:10 UTC 📊 Assets: 16 files 📝 What's New: <div dir="rtl"> • بهینه‌سازی Full mode Apps Script: batch کردن Edge-DNS cache lookupها (PR #958 by @dazzling-no-more). مسیر CodeFull.gs قبلاً برای هر DNS query داخل batch یک CacheService.get جداگانه می‌زد. روی batchهای شلوغ، همین چند round-trip داخلی به CacheService می‌تونست latency بسازه، مخصوصاً وقتی چند lookup مشابه... 🔗 View on GitHub 📥 Download Release

🚀 New Release Alert! 📦 MITM-DomainFronting 🏷 Category: Community Suggested 🔖 Version: MITM-DomainFronting v17 📅 Released: 2026-05-15 14:18 UTC 📊 Assets: 0 files 📝 What's New: fix github and fastly (reddit, ...) 🔗 View on GitHub 📥 Download Release