Tennisbot (v7)：多会话多 Agent 个人助手的架构与开发日志
发表于 - Posted on
字数 - Word count:
3.3k
阅读时间 - Reading time ≈
12 mins.
Tennisbot v7 是一个基于 OpenAI Agents SDK 的多会话、多 Agent 个人助手。本文从入口与目录结构讲起，沿着 WebUI/CLI 两条执行链路下钻到会话存储、handoff、工具与多模态输入，记录它如何在“可自修”的约束下迭代到可用形态，并给出关键设计取舍与踩坑复盘。
使用codex进行Vibe-Coding的过程记录 - Process Log of Vibe-Coding with Codex
发表于 - Posted on
编辑于 - Edited on
字数 - Word count:
1.1k
阅读时间 - Reading time ≈
4 mins.
干涉花纹
发表于 - Posted on
编辑于 - Edited on
字数 - Word count:
115
阅读时间 - Reading time ≈
1 mins.
夏末
发表于 - Posted on
编辑于 - Edited on
字数 - Word count:
104
阅读时间 - Reading time ≈
1 mins.
野湖边的野草
发表于 - Posted on
编辑于 - Edited on
字数 - Word count:
1k
阅读时间 - Reading time ≈
4 mins.
关于埃德蒙顿的野外湖边的一些常见的野草的记录
LLM排行榜及测评：25/08/17 - LLMs Leaderboard and Evaluation：25/08/17
发表于 - Posted on
编辑于 - Edited on
系列 - Series
LLM排行榜 - LLM Leaderboard
字数 - Word count:
2.6k
阅读时间 - Reading time ≈
9 mins.
本表格汇总了常用大语言模型在常用评测榜单上的表现。榜单涵盖人类偏好、知识与推理能力、数学能力、代码能力、多模态能力等多个方面。
This table summarizes the performance of popular large language models across well-known benchmark leaderboards. These rankings cover a range of capabilities, including human preference, knowledge and reasoning, mathematical skills, coding ability, and multimodal performance.
This table summarizes the performance of popular large language models across well-known benchmark leaderboards. These rankings cover a range of capabilities, including human preference, knowledge and reasoning, mathematical skills, coding ability, and multimodal performance.
文章精读 - Paper Reading 2：Machine learning potentials for metal-organic frameworks using an incremental learning approach
发表于 - Posted on
编辑于 - Edited on
字数 - Word count:
2.4k
阅读时间 - Reading time ≈
9 mins.
哥德尔不完备性定理
发表于 - Posted on
编辑于 - Edited on
字数 - Word count:
4.1k
阅读时间 - Reading time ≈
15 mins.
尝试用通俗易懂的方法证明哥德尔不完备性定理
使用Apptainer构建NequIP容器，并在DRAC上运行
发表于 - Posted on
编辑于 - Edited on
字数 - Word count:
3.7k
阅读时间 - Reading time ≈
13 mins.