Nov 22, 2026On-premise AI Lab🧪 HF trending 683 GGUF multimodal small model, 0.5B runs voice chat, Gemma 4 fits into 8GB via PLE | On-device AI LabThe highlight of this issue is Unsloh’s release of the Inkling‑2‑Mini GGUF version, with a Hugging Face heat score of 683 and over 800,000 downloads, a single 3.5 GB file can run image‑text QA on 8 GB VRAM or Mac unified memory; Alibaba’s InclusionAI launches the 0.5B‑parameter Ming‑3.0‑audio‑flash in ONNX format, running end‑to‑end voice dialogue directly on CPU with official latency claimed under 500 ms; Go
Nov 21, 2026Tech News🎙️ Anthropic's Record-Breaking IPO, Claude Claims to Have Solved the Riemann Hypothesis? | AI Daily PodcastAnthropic debuted on Nasdaq at a $107.3 billion valuation, surging 17% on its first day to set a new AI‑startup IPO record; I’m Muyan, welcome to Mark’s Tech Insights, where we’ll cover this blockbuster IPO, claims that Claude helped prove the 160‑year‑old Riemann hypothesis, and progress on China’s new token standard, and why these three developments could rewrite the rules of the AI industry.