2026-10-24 AI 日报

开源项目

1. Zagitta/gpu-unlock

   Unlock FP/BF16 instruction issue rate on Nvidia 3000's cards to their professional equivalents

2. makecindy/cindy

   Consider it done. The open-source AI agent that works out of the box · 想到，就能做到。开源、开箱即用的 AI Agent。

3. aws/agent-toolkit-for-aws

   Official, AWS-supported MCP servers, skills, and plugins to help AI agents build on AWS

4. google-gemini/gemini-cli

   An open-source AI agent that brings the power of Gemini directly into your terminal.

5. Nutlope/hallmark

   Anti-AI-slop design skill for Claude Code, Cursor, and Codex.

6. AgriciDaniel/claude-seo

   Universal SEO skill for Claude Code. 26 sub-skills + 19 sub-agents covering technical SEO, E-E-A-T, schema, GEO/AEO, agent readiness (Lighthouse Agentic…

7. nightly-labs/openbot

   A local-first desktop workspace for persistent AI teammates. Run Codex, Claude, and Grok with dedicated workspaces, task queues, file sharing, browser…

8. luongnv89/claude-howto

   A visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.

9. AstrBotDevs/AstrBot

   AI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨

10. HKUDS/DeepTutor

    DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.

论文

本统计日暂无条目。

行业新闻

1. GPU-unlock: FP/BF16 instruction issue rate on Nvidia 3000's cards

2. Cerebras Deploys Trillion-Parameter AI Models

3. Why is AI generated language so hard to understand?

   arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that…

4. Five Qwen3.8-27B models match Fable 5 Using GVS5H Harness

5. Oai-math-revised: a community hub for revising OpenAI's math write-ups

6. Making My Blog Readable by AI Agents

7. Argus – a free, open-source agentic security operations platform – MIT

8. Claude Code slyly adding users' emails to outbound HTTP

9. Show HN: Enzyme – a programming language that compiles wikis for agents

10. Show HN: Argus- open source security platform to protect against agentic attacks

社区动态

1. Person who is reverse engineering his tv gets banned from OpenAI You aren't allowed to stop your tv from spying on…

   Person who is reverse engineering his tv gets banned from OpenAI You aren't allowed to stop your tv from spying on your family. You have to ask for permission…

2. OpenAI and Anthropic blocked the reverse engineering of smart TVs spying on us

   REA (Reserve Engineer Anything) was blocked by Claude one day later... https://x.com/AbhinavXJ/status/2113590263373508704 Time for the pebbles to own the…

3. memt-1.0 architecture breakdown gated deltanet-2, mixture-of-storage, and 66x smaller storage footprint for agent memory [D]

   been digesting the details behind memt-1.0's memory-native architecture recently, and the engineering tradeoffs between full context kv caching and…

4. Kimi K3.1 removes cyberattack training data from open weight release

   This is due to China's new export restrictions primarily but also the growing caution around cybersecurity exploits popping up.

5. A 4B coding model that can generate 3D scenes

   I’m sharing an early version of a 4B model I’m training for general coding and 3D scene generation. Small models have been pretty rough at 3D in my tests.…

6. Meta trained small AI models to read raw bytes instead of tokens. They start out worse, but the paper predicts they end up better

   https://arxiv.org/pdf/2609.12303 this paper suggests a model that skips models reading tokens that and has the model read raw bytes, basically one letter at a…

7. 8 GB GPU, 262K context 3.5 BPW Qwen-Flash-Next Coder 70 T/s using llama.cpp

   The methodology Qwen-Flash-Next has n-gram, MOE packing above its weight but it's still 6B active params. We exploit this by keeping most MOEs in RAM…

8. I took apart Claude Code, OpenCode and Pi, then turned what I learned into a free course on harness engineering

   The fastest way to learn harness engineering? Build Claude Code from scratch. Why harness engineering? LangChain kept the same model and changed only the…

9. ICLR 2027 is still handing out desk rejects [N]

   Thank you for your efforts in protecting the community from AI slop.

10. Do benchmarks even represent the real life environment workflow?

    Don't get me wrong, I am aware that latest benchmarks do represent the actual workflow much better than the previous benchmarks. But, when I was seeing the…

官方发布

1. Z.ai GLM 5.3 Flash (zai-glm-5-3-flash) is now available.

   Z.ai GLM 5.3 Flash (zai-glm-5-3-flash) is now available. A multimodal 320B-parameter (18B active) Mixture-of-Experts model from Z.ai with a 1M-token context…

2. Kimi K3.1 Pro

3. The Decisions API is now generally available with gpt-6-luna.

   The Decisions API is now generally available with gpt-6-luna. Cache shared input across requests for prompt-cache discounts, include images in individual…

4. MiniMax M3.1 Flash

   An ultra-efficient open-weight model for everyday work.

5. Securing our frontier model training environment

   OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.

6. Using thousands of AI agents to uncover new leads in biological data

   See how OpenAI, PerturbAI, and Scripps Research used distributed AI agents to uncover promising research leads in data from millions of brain cells.

7. Streaming comes to API Gateway: real-time and AI-ready APIs, without changing your spec

   API Gateway now supports real-time streaming—including Server-Sent Events, WebSockets, gRPC, and chunked transfer encoding—in Public Preview without requiring…

8. Voxtral 3

   Voxtral 3: Mistral's voice agent SDK for real-time speech in 50+ languages — accurate transcription, emotional TTS, deep tool use. EU-hosted or self-deployed.

9. Fast mode is available for Claude Opus 5.5 on Claude Models on Google Cloud (Anthropic managed).

   Fast mode is available for Claude Opus 5.5 on Claude Models on Google Cloud (Anthropic managed).

10. We've launched Claude Models on Google Cloud (Anthropic managed), the Claude API operated by Anthropic and reached…

    We've launched Claude Models on Google Cloud (Anthropic managed), the Claude API operated by Anthropic and reached through your Google Cloud project. Google…