Mailing list archive

• llama.cpp goes native MCP, DeepSeek V4 & TTFT gains Fri, 04 Sep 2026
• Meta's Muse-Glimmer, Nemotron MoE, and Local A2A Fri, 14 Aug 2026
• Liquid LFM2.5 and non-transformer architecture Fri, 07 Aug 2026
• Kimi K3 and the scaling limits of local hardware Fri, 31 Jul 2026
• Kimi K3 open weights and the shift to sparse MoEs Fri, 24 Jul 2026
• Running 27B models on a phone: the rise of 1-bit LLMs Fri, 17 Jul 2026
• Ollama just raised $65M. Local AI is real now. Fri, 10 Jul 2026

Subscribe