本地AI报告

作者： Jakub Rusinowski · 最后更新： 2026年8月31日

关于本地LLM现状的双周报告：新模型发布及显存需求、GPU价格变动和工具更新。有日期、可引用的期刊。

• Local AI Report #4 — Best Laptops for Local AI · 2026年8月31日 — In a laptop, memory bandwidth and memory capacity sit at opposite ends of the price list. The fastest memory you can buy comes in the smallest quantity, the largest comes on the slowest bus, and exactly one part escapes the trade-off.
• Local AI Report #3 — Best Small LLMs for 8 GB and 16 GB RAM Laptops · 2026年8月25日 — An 8 GB laptop leaves about 6.4 GB for a model, and a 16 GB laptop about 12.8 GB. Here is what actually fits, how much context you get, and how fast it runs with no discrete GPU.
• 第 #2 — Local AI Report #2 — The Best Open-Source Coding Models Right Now · 2026年7月8日 — A field guide to the strongest open-weight coding models in mid-2026: the SWE-bench frontier (DeepSeek V4-Pro, GLM-5.2, Kimi K2.6), the best coder you can actually download (Qwen3-Coder), and the setup that gives the most code per dollar on a single GPU.
• 第 #1 — Local AI Report #1 — The Mid-2026 Local LLM & Hardware Landscape · 2026年7月7日 — The first biweekly digest on the state of local LLMs: the mid-2026 model generation (Gemma 4, Qwen 3.6, DeepSeek V4), where GPU prices actually stand, and what the current best-value setup looks like.