Feeds

• architecture 1

• hardware 4

• open source 4

Recent

• 2026-10-24 A document platform for humans and agents: the architecture
• 2026-10-22 FP4 KV cache on the local fleet: 1.6x the pool, benchmarks pending
• 2026-10-22 Does MoE on vLLM make sense? Qwen3.8-Flash-Next vs dense Qwen3.8-27B on two DGX Sparks
• 2026-10-17 Private inference on RTX 5090 - how much saved?
• 2026-10-15 Quantum computers you can buy
• 2026-10-14 Web scrapers for local LLMs
• 2026-10-13 Deep-research subagents bottleneck on prefill — how my scraper fixes it
• 2026-10-13 Web search for local LLMs: SearXNG vs native Claude vs hosted search APIs
• 2026-08-02 broker_parrot — a self-hosted job fleet on the database you already run