29 Janvier 2026 Niamey 🇳🇪 est actuellement attaquée les affrontements entre L’ARMÉE digne du NIGER 🇳🇪 et les terroristes chapeautés par la France 🇫🇷 essaient de prendre le contrôle de l’aéroport de Niamey L’hiver Noir est donc déclaré

#bikers

a while ago #bond

live drills ongoing

#PUT options virtual vapor looking good

💣💥 CENTCOM looks ready 💣💥 LOADED weekend fireworks - $$$ PRICED-IN ? live drills ongoing 38+hrs 💥💥💥massive build up 💥💥💥 reconnaissance ⌯✈︎ 🚤 🛰️ super active both sides

Moltbot is an open-source personal AI assistant that can be integrated with various messaging platforms and can perform tasks like managing calendars and emails. It can be powered by different language models, including those from AirLLM, allowing for customizable interactions and functionalities. or route to other models using small SATS BTC-LN routstr. otrta.me gpt-40-mini or others. K2.5 has no roustr yet - running locally need at least 24GB VRAM GPU + 800GB SSD 16GB RAM PC / VPS / TEMPRENT container

#moltbot is AI agent - run in own small PC now new option emerges #moltworker who do not have HW or VPS or other Rental containers learn n absorb architcture -> API openrouter pay be use keys using BTC-LN sats - few providers in nostr for backend models or run own small K2.5 or deepseek or any other paid ones

FOSS AI

You can now run 70B model on a single 4GB GPU and it even scales up to the colossal Llama 3.1 405B on just 8GB of VRAM. #AirLLM uses "Layer-wise Inference." Instead of loading the whole model, it loads, computes, and flushes one layer at a time - No quantization needed by default - Supports Llama, Qwen, and Mistral - Works on Linux, Windows, and macOS 100% Open Source.