Briefing · 2026-10-20

技术简报 - 2026-10-20

6 immediate1 worth tracking8 sources
On this page6 sections

The briefing's title, the editors' summary and their notes are written in Chinese, as are skill and knowledge entries.

Overview

#

这一期的信号指向同一件事:模型的竞争点正从单项能力转向智能体任务的完成度和单位成本。先看 Google DeepMind 的 Gemini 4 Argon:当天通过 API 全面开放,每百万输入 token 2 美元、输出 10 美元。开放权重阵营同期跟上,Meta 的 Muse Spark Open(816B 多模态 MoE)和 Reflection 的 Beam(501B,Apache 2.0)都得先算显存;DeepSeek 的 V4.1 Flash 新版则主打更强也更省。训练与部署栈也在动:Agent Lightning 让线上 harness 直接参与强化学习,vLLM v0.32.0 把 GLM-5.x 在 SM100 上的 KV 缓存默认改成 FP8,升级前要实测。

Immediate attention

#

The highest-rated published signals; the editors' pinned items come first.

Worth tracking

#

Worth following, but each needs more context or confirmation before it becomes a priority.

Sources

#

The public sources behind the published signals in this briefing.

Google DeepMind

Google DeepMind Blog - 2026-10-14

Open source

Meta

Hugging Face Blog - 2026-10-15

Open source

Microsoft Research

Microsoft Research Blog - 2026-10-08

Open source

vLLM

vLLM Releases - 2026-10-18

Open source

DeepSeek

量子位 - 2026-10-17

Open source

Reflection

Hugging Face Blog - 2026-10-16

Open source

Google Research

InfoQ 中文 - 2026-10-20

Open source

Google AI Blog

Public source

Subscribe

The public feeds carry published briefings only, never drafts or archived ones. They are in Chinese.