与「SCHEDULER」相关的搜索结果

开源项目 LoopX：超长程 Agent 自主运行 200+ hours，状态不漂移。 我的技术主张是：LLM 上下文有限，长程 Agent 需要外置状态，通过完备的状态管理、监督和规划，让 Agent 无人干预时跑得稳、持续有产出；有人干预时跑得更好，能吸收反馈继续演进。 两条真实 trajectory 分别跨越 220.7 / 272.9 小时，跨多轮执行、等待、人工决策、writeback 与 resume 后，整个 loop 仍能找回目标、证据和下一步。 目前 LoopX 已有 3 个 showcase：auto PR issue fix、AutoML experiment 和 auto coredump fix。 以 OpenViking 开源仓库的 PR issue fix 为例，Agent 不只是循环写代码。它需要持续理解 issue 的不同状态，判断何时开发、何时等待、何时请求 review，处理 CI、冲突和上游变化，并连续交付多个 PR。 这对应 LoopX 的 domain state 管理：领域系统决定真实状态，LoopX 负责把状态投影成下一步可执行的工作。 与此同时，Agent 还可以在干活过程中实现能力自进化。当它发现现有系统缺少某项能力时，可以提出 feature、完成开发与验证、发布新的离线或在线版本，再使用新能力继续原来的任务。 长程 Agent 天然适合自进化，“完成工作”和“升级完成工作的系统”可以在同一条长程轨迹中发生。 LoopX 把这些信息外置成结构化控制面： • Goal / Vision：目标是什么，什么不能被局部优化牺牲 • Todo / Gate：当前执行的 frontier，以及必须留给人的关键判断 • Identity / Authority：谁能 claim、writeback、approve • Evidence / Receipt：每次推进留下什么可回读证据 • Quota / Scheduler：何时继续执行，何时安静等待 • Handoff / Recovery：换模型、换会话、换 host 后如何恢复 你也可以把它理解为一块专门给 Agent 设计的可执行 Kanban。 普通看板只展示“谁在做什么”；LoopX 的状态会直接约束和驱动下一次 bounded turn，让看板本身成为执行系统的一部分。 这套系统最强的地方是通用性。它不只可以修 PR，还可以做 auto research、长期实验、自媒体运营、复杂 feature 开发和办公任务。 Agent 不再只是一次性的回答机器，而可以围绕一个人的 vision，长期工作、等待、吸收反馈、积累证据并持续演进。 LoopX 从一开始 build in public：状态协议、CLI、控制面实现和真实运行轨迹都进入了开源仓库；它也已经和 OpenViking、NoKV 等 agent infra 项目形成了开源合作伙伴关系。 我希望 LoopX 最终能放大每个人的 vision 和想象力。只要你有自己的目标、想法或技术主张，就能拥有一个全天候继续工作和探索的 Agent 系统，帮助你把愿景一点点变成现实。 欢迎试用、提 issue、贡献代码，或者用一个真实的 multi-day task 跑 LoopX。

xAI has been shipping Grok Build updates non-stop If you have not been keeping track, here is what xAI has rolled out up to v0.2.11: New commands & core features → Integrated 𝕏 search and much faster web search → Added /export, /login, /usage, and /config-agents → Added an interactive read-file viewer with PowerPoint text extraction → Added Always-approve mode to streamline permissions → Auto-installed shell completions for bash, zsh, and fish Expanded platform support → Added Windows ARM64 and macOS x86_64 support → Improved terminal support across Warp, VTE-based terminals, JetBrains, and legacy Windows Console → Fixed copy/paste issues across Linux Wayland, WSL, and Windows Agent & context improvements → Subagents now share the terminal backend, scheduler, and monitor across sessions → Added proactive system reminders and laziness detectors to keep the model on task → Improved context compaction and memory usage for chat history → Auto-backgrounds long-running bash-mode commands triggered via ! UX, media & UI upgrades → Boosted terminal video playback to 30fps → Added multi-image paste, drag-and-drop, and temporary macOS screenshots → Added instant loading indicators when switching models → Improved clickable markdown links and rendered links inside table cells → Smoother plan-mode controls and absolute line numbers in the edit panel Stability & fixes → Increased default retry budget to ~5 minutes → Increased Unix ulimits to prevent crashes → Hardened background tools to handle timeouts and self-kills → Fixed rendering bugs across UTF-8 output, large monitors, Windows contrast, and more xAI is moving insanely fast Grok Build is going from early CLI to a serious agentic coding environment very quickly

Bug fixes shipping to Grok Build 0.2.7 (release notes will be available in the TUI) • Wrap around on Up/Down in slash menu and history search • Restore subagent UI and replay on session resume • Fix Ctrl+Delete on windows • Windows drag-and-drop screenshot images + Ctrl/Alt+V image input • Add Windows-friendly alternatives for Ctrl+Enter / Ctrl+; • Strip base64 str images from read_file tool calls & pass as multimodal vision tokens • Share terminal backend, scheduler, monitor across subagent sessions • Image pipeline improvements (fix truncation, min pixels) • /login command to reauth from inside TUI • Add /usage • Handle Cmd+A in prompt to select all text in ghostty

4-Vulnerability Exploit Chain in DataEase My team found a 4-vulnerability exploit chain allowing unauthenticated RCE on DataEase. Combined with a previously published vulnerability (CVE-2026-23958) - these new vulns complete the attack chain, bypassing JDBC, SQL Injection and a Quartz scheduler injection that runs periodically and executes a crafted payload on the machine. We have also a video showing the exploit POC in action :) Read the full blog:

Stop guessing which models fit in your VRAM! llmfit is a CLI tool that auto-detects your hardware and ranks 206 models by what actually runs on your system. You download a 70B model and hope it fits. Or you estimate memory requirements across quantization levels and still end up with models that crash or run too slow. llmfit changes that. It detects your CPU, RAM, GPU, and VRAM, then scores every model in its database against your hardware. Instead of assuming one quantization level, it tries the best quality that fits. Starts with Q8_0, walks down to Q2_K if needed. If nothing fits at full context, it tries half context. You get the highest quality model that actually works. Each model gets scored on Quality, Speed, Context, and Capability. The weights shift based on what you're doing. Chat models prioritize speed, reasoning models prioritize quality. Run it as an interactive TUI to browse models, use CLI mode for a quick table, or get JSON output for scripts. There's a REST API for cluster schedulers. You can also run it in reverse. Give it a model you want to run and target performance, it tells you what hardware you need. The real value: you see ranked options before downloading anything. No more burning bandwidth on 50GB models that won't run. It's 100% open source. Link to llmfit in comments!

I.O.I 3rd MINI ALBUM [I.O.I : LOOP] 𝐏𝐫𝐨𝐦𝐨𝐭𝐢𝐨𝐧 𝐒𝐜𝐡𝐞𝐝𝐮𝐥𝐞𝐫 ♾️ 2026. 05. 19 TUE 6PM (KST) #아이오아이# #IOI# #IOI_LOOP#