AI Resonance · 阅读最新日报 · AI 入门推荐

2026-09-21 AI 日报

统计日时区：America/Los_Angeles · 已结算

google/ax · Nasiko-Labs/nasiko · BuilderIO/agent-native

在应用中阅读本期

开源项目

  1. google/ax

    Google's open agentic orchestration runtime

  2. Nasiko-Labs/nasiko

    Developer Control Plane for your AI Agents

  3. BuilderIO/agent-native

    A framework for building agentic apps

  4. akitaonrails/ai-memory

    Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors

  5. stablyai/orca

    Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and remote runtime.

  6. TNT-Likely/PanWatch

    盯盘侠 PanWatch · 自托管 AI 盯盘助手，集成 TradingAgents 多 Agent 投资决策 | A股/港股/美股实时监控、持仓管理、智能分析、全渠道推送

  7. dream-num/univer

    The Office Harness for AI Agents — Spreadsheets, Docs, Slides, Canvas, Relational Tables, and PDF in one runtime.

  8. weave-os/router

    Model router for agentic systems. Routes every prompt to the right model in <50ms. Cut costs 40-70% with just an endpoint change.

  9. affaan-m/ECC

    The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode,…

  10. yynxxxxx/Codex-X

    OpenAI Codex 桌面端/CLI 的可视化管理工具，具有Provider/API 切换、会话同步、提示词注入、Skills/MCP 管理、TOML 配置可视化的跨平台工具。

论文

  1. EvoOntology: A Self-Evolving Ontology Layer for Data Agents

    Data agents aim to fulfill natural-language instructions over heterogeneous data, including tables, files, and databases. However, data agents face a…

  2. Grounded Skill Synthesis from Code at Scale for Agentic Intelligence

    Reusable skills give agents transferable procedural knowledge, making scalable acquisition essential for extending agents beyond prior experience. Existing…

  3. RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents

    Computer-use agents (CUAs) have advanced along two separate lines: graphical interaction and software development through code and the command line. Real…

  4. CodeMidas: Scaling Agentic Coding RL Environments from Code Itself

    Training capable coding agents via reinforcement learning (RL) requires diverse tasks with reliable verifiers. Open-source codebases offer a rich source of…

  5. OmniVChat: Synthesizing, Benchmarking, and Training for Native Audio-Visual Dialogue

    We define OmniVChat (Omni Video Chat) as the task of native audio-visual dialogue between a user and an omni model. In OmniVChat, omni models directly and…

  6. IntBMoE: Integrating Block-Level Conditioning into Expert Composition for Full-Participation Mixture-of-Experts

    Mixture-of-Experts (MoE) scales capacity, but existing designs cannot set three quantities independently. For a single token, participation is how many…

  7. Paint-Anything: Unified Any-Color Control for Image Generation and Editing

    Professional design requires any-color control: the ability to specify an object's target color with any 24-bit hex value for image generation and editing.…

  8. Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design

    Professional graphic design is a long-horizon agentic task in which structured, editable artifacts emerge from many interdependent actions, yet outcomes admit…

  9. BI-Agent and BI-Bench: Towards Automating End-to-End Business Intelligence

    Business intelligence (BI) is a cornerstone of enterprise decision-making and is widely used by enterprise users in software such as Power BI and Tableau. In…

  10. OmniVBench: A Benchmark and Large-Scale Dataset for Omni Reference-to-Video Generation

    Reference-to-video (R2V) generation is evolving toward increasingly general and versatile reference control, giving rise to the emerging paradigm of omni R2V…

行业新闻

  1. Grok 4.7

    SpaceXAI's most powerful model for coding and knowledge work. Twice as fast, at half the price of comparable models.

  2. Kev: Tiny Jev-like family of decision models built on top of Qwen3.5

  3. Can gzip be a language model?

  4. Transformers Explained Visually

  5. M5 Ultra Mac Studio Review: The Dream Mac for Local AI Agents

  6. Advisory Group on Mathematics and Artificial Intelligence

  7. Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com

  8. Claude Status – Elevated errors for multiple models

  9. macOS 27: Workaround to avoid downloading AI models and save storage

  10. The Claude Delusion

社区动态

  1. Grok 4.7 is here.

    Grok 4.7 is here. It's a notable improvement over Grok 4.6 at the same price and speed. https://t.co/H3OTBbXyvO

  2. We’re working with an independent advisory group of mathematicians to help OpenAI responsibly share advances in AI and…

    We’re working with an independent advisory group of mathematicians to help OpenAI responsibly share advances in AI and mathematics. The group will advise on…

  3. Kimi K3 is now on Amazon Bedrock!

    Kimi K3 is now on Amazon Bedrock! Run coding, document analysis, and extended agent workflows with Bedrock's access, encryption, and auditing controls.…

  4. OpenAI solved 100 open problems in math

  5. M5 Ultra Mac Studio Review: The Dream Mac for Local AI Agents - MacStories

  6. Introducing Grok 4.7

  7. Kev: tiny Jev-like decision models (0.8B/4B/9B) on Qwen3.5 you can train and run locally - the 9B fits a 32GB Mac

    Been playing with judge models in my eval pipeline, so this landed at the right time. Kev is a small family of Jev-architecture decision models (0.8B, 4B, 9B)…

  8. [R] Complex KDA: Understanding and Enhancing the Expressivity of Kimi Delta Attention

    Github: https://github.com/OpenEuroLLM/ComplexKDA HuggingFace: https://huggingface.co/collections/openeurollm/complexkda Arxiv:…

  9. These Were NOT Rogue AI Escapes. Just SLOPPY Firewall Failures. [N]

    The headlines right now are full of stories about AI models "escaping their sandboxes" and literally killing all humans, lol. I've even heard several…

  10. 16GB (and in many cases 12GB) is the max vram most people will ever reasonably have

    This sub is, needless to say very niche and skewed towards the high end. There are tons of extremely high end setups here with multiple gpu's etc. Even 24GB…

官方发布

  1. Introducing Grok 4.7

  2. Qwen-Image-2.1: Compact, Efficient, and Unified Image Creation

    We are excited to open-source Qwen-Image-2.1, an image model in the Qwen family that balances generation quality, inference efficiency, and cost.…

  3. Introducing Grok Voice Transcribe 2.0

  4. Qwen/Qwen-Image-2.1-PE-T2I

    text-to-image

  5. Privacy Center in ChatGPT

    Privacy Center is rolling out to signed-in ChatGPT Free, Go, Plus, and Pro users. It brings together information about chat privacy, memory, personalization,…

  6. Credit scores in Finances

    You can now track your credit score in ChatGPT. Securely connect your Experian® credit report and VantageScore® 3.0 credit score to get personalized insights…

  7. Advisory Group on Mathematics and Artificial Intelligence

    OpenAI is working with an independent Advisory Group on Mathematics and Artificial Intelligence to guide the review and communication of emerging AI results.

  8. Introducing Gemini 3.8 Live and 3.8 Live Extended Thinking

    Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking are our most advanced live dialogue models yet, built for natural conversation.

  9. Qwen3.8-LiveTranslate: Names the speaker. Carries the meaning.

    Simultaneous interpretation is not only about translating fast — it must also hear clearly and translate accurately. Qwen3.8-LiveTranslate rebuilds real-time…

  10. Partnering with Accenture on embedded evaluation