GLM-5:从“氛围编码”到自主工程,1M上下文开源模型 | GLM-5
> 📌 **项目地址**:[zai-org/GLM-5](https://github.com/zai-org/GLM-5) | ⭐ 4,013 颗星 | 📜 未标注 ## 项目现状:三个版本,仅开放API z.ai 开了一个 GitHub...
> 📌 **项目地址**:[zai-org/GLM-5](https://github.com/zai-org/GLM-5) | ⭐ 4,013 颗星 | 📜 未标注 ## 项目现状:三个版本,仅开放API z.ai 开了一个 GitHub...
📌 项目地址:alexzhang13/rlm | ⭐ 4,855 颗星 | 🔧 Python | 📜 未标注 超长上下文的老问题与新解法 当前大语言模型(LLM)在上下文长度上有硬性天花板——无论是4K、128K还是1M token,总有一...
📌 项目地址:calesthio/OpenMontage | ⭐ 5,130 颗星 | 🔧 Python | 📜 未标注 核心价值:让AI Coding Assistant变成视频制作工作室 OpenMontage 宣称自己是“全球首个开源...
> 📌 **项目地址**:[continuedev/continue](https://github.com/continuedev/continue) | ⭐ 33,838 颗星 | 🔧 TypeScript | 📜 未标注 ## 项目现...
> 📌 **项目地址**:[google-research/timesfm](https://github.com/google-research/timesfm) | ⭐ 21,714 颗星 | 🔧 Python | 📜 未标注 ## 这...
📌 项目地址:DeusData/codebase-memory-mcp | ⭐ 4,801 颗星 | 🔧 C | 📜 未标注 它是谁?解决了什么问题? 如果你在用 AI 编码助手(如 Cursor、Copilot、Claude Code)处...
📌 项目地址:mikeroyal/Self-Hosting-Guide | ⭐ 20,819 颗星 | 🔧 Dockerfile | 📜 未标注 这个项目解决了什么问题? 自托管(Self-Hosting)意味着你在自己的硬件或私有服务器上...
📌 项目地址:trycua/cua | ⭐ 18,088 颗星 | 🔧 HTML | 📜 未标注 两个核心模块,一个共同目标 cua 项目为“计算机使用智能体”(Computer-Use Agent)提供完整的基础设施。它由两个层次构成: ...
> 📌 **项目地址**:[teslamate-org/teslamate](https://github.com/teslamate-org/teslamate) | ⭐ 8,191 颗星 | 🔧 Elixir | 📜 未标注 ## 为什...
📌 项目地址:huggingface/OpenEnv | ⭐ 2,206 颗星 | 🔧 Python | 📜 未标注 它要解决什么问题? 在训练基于强化学习的智能体(特别是大语言模型的后训练)时,环境的构建和交互往往是开发中的瓶颈。传统的 ...
> 📌 **项目地址**:[GorvGoyl/Clone-Wars](https://github.com/GorvGoyl/Clone-Wars) | ⭐ 35,269 颗星 | 📜 未标注 ## 这个项目只做一件事 把 GitHub 上...
📌 项目地址:LMCache/LMCache | ⭐ 8,575 颗星 | 🔧 Python | 📜 未标注 为什么要关注这个项目 大模型推理中,长上下文(如整本书、多轮对话)的首Token延迟(TTFT)一直是性能瓶颈。传统方案每次请求都...