addyosmani/agent-skills
Shell · ★ 56,831 · 🍴 6,126 · 📈 2,656 stars today
Production-grade engineering skills for AI coding agents.
中文介绍 为AI编码智能体提供的生产级工程技能集合，封装了常见工程实践与流程，使agent能更可靠地参与软件开发。适用于利用AI agent进行代码生成、测试、重构等任务。
2026-06-13
← 历史归档music-assistant/server
Python · ★ 1,786 · 🍴 422 · 📈 20 stars today
Music Assistant is a free, opensource Media library manager that connects to your streaming services and a wide range of connected speakers. The server is the beating heart, the core of Music Assistant and must run on an always-on device like a Raspberry Pi, a NAS or an Intel NUC or alike.
中文介绍 开源媒体库管理器Music Assistant的核心服务层，连接各类流媒体服务与智能音箱，提供统一的音乐播放与元数据管理。适合家庭多设备场景下的集中式音乐控制。
mattermost/mattermost
TypeScript · ★ 37,634 · 🍴 8,713 · 📈 388 stars today
Mattermost is an open source platform for secure collaboration across the entire software development lifecycle..
中文介绍 面向软件开发生命周期的开源协作平台，支持即时消息、文件共享、渠道组织与集成。适用于对安全性和自托管有要求的团队，可作为Slack的替代方案。
apple/container
Swift · ★ 35,129 · 🍴 980 · 📈 3,504 stars today
A tool for creating and running Linux containers using lightweight virtual machines on a Mac. It is written in Swift, and optimized for Apple silicon.
中文介绍 在Mac上通过轻量级虚拟机创建与运行Linux容器的工具，使用Swift编写，针对Apple Silicon深度优化。适用于需要在macOS下高效测试或部署Linux应用的开发者。
iptv-org/iptv
TypeScript · ★ 118,045 · 🍴 6,303 · 📈 179 stars today
Collection of publicly available IPTV channels from all over the world
中文介绍 全球公开IPTV频道的集合仓库，整理了大量可播放的电视直播源，涵盖各国频道。用户可通过播放器直接加载M3U列表，适合希望免费观看世界各地电视节目的人群。
obra/superpowers
Shell · ★ 226,017 · 🍴 20,089 · 📈 1,275 stars today
An agentic skills framework & software development methodology that works.
中文介绍 一套智能体技能框架与软件开发方法，强调实用性和可操作性。帮助开发者构建具备自主能力的AI代理，适用于需要将LLM落地到具体编程或管理流程的场景。
refactoringhq/tolaria
TypeScript · ★ 15,777 · 🍴 1,077 · 📈 369 stars today
Desktop app to manage markdown knowledge bases
中文介绍 桌面端Markdown知识库管理应用，提供笔记组织、搜索与版本控制功能。适合习惯以纯文本形式管理个人知识体系的技术人员或知识工作者。
maziyarpanahi/openmed
Python · ★ 3,197 · 🍴 306 · 📈 515 stars today
open-source healthcare ai
中文介绍 开源医疗AI项目，涵盖临床自然语言处理、医学影像分析等方向。旨在为医疗研究人员与开发者提供可复用的模型和工具，降低AI在医疗领域的应用门槛。
LMCache/LMCache
Python · ★ 8,642 · 🍴 1,288 · 📈 28 stars today
LMCache: Supercharge Your LLM with the Fastest KV Cache Layer
中文介绍 为LLM推理提供超高速KV缓存层，通过优化缓存存储与检索大幅降低延迟。适用于需要频繁复用历史上下文的长对话、文档分析等场景，提升大模型的响应速度。
phuryn/pm-skills
★ 16,993 · 🍴 1,742 · 📈 827 stars today
PM Skills Marketplace: 100+ agentic skills, commands, and plugins — from discovery to strategy, execution, launch, and growth.
中文介绍 产品经理技能市场，包含100多个面向AI agent的技能、命令和插件，覆盖从需求发现到增长运营的全流程。适合使用AI辅助产品管理工作的PM或构建自动化agent的开发者。
masterking32/MasterDnsVPN
Go · ★ 6,017 · 🍴 540 · 📈 400 stars today
Advanced DNS tunneling VPN for censorship bypass, optimized beyond DNSTT and SlipStream with low-overhead ARQ, resolver load balancing, high packet-loss stability and speed.
中文介绍 基于DNS隧道的进阶VPN工具，通过低开销ARQ、解析器负载均衡和抗丢包技术突破网络审查。优化了传统DNSTT和SlipStream，在恶劣网络环境下仍保持较高速度与稳定性。
msitarzewski/agency-agents
Shell · ★ 112,430 · 🍴 18,334 · 📈 1,026 stars today
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
中文介绍 一站式AI代理机构框架，内置前端设计师、Reddit运营、趣味内容注入等多种专业化agent，每个agent都有独立人格和工作流程。适用于需要自动化完成多角色协作任务的个人或小团队。
microsoft/PowerToys
C · ★ 134,339 · 🍴 8,063 · 📈 103 stars today
Microsoft PowerToys is a collection of utilities that supercharge productivity and customization on Windows
中文介绍 微软官方推出的Windows系统增强工具集，包含颜色选择器、窗口管理器、批量重命名等实用功能。旨在提升操作效率与系统自定义能力，适合所有Windows用户。
EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
👍 104
Large language model (LLM) agents have achieved strong performance on a wide range of benchmarks, yet most evaluations assume static environments. In contrast, real-world deployment is inherently dynamic, requiring agents to continually align their knowledge, skills, and behavior with changing envir
中文介绍 EvoArena通过追踪记忆演化，使LLM Agent在动态环境中持续对齐知识、技能与行为，解决静态评估与真实部署的差距，提升鲁棒性。
MiniMax Sparse Attention
👍 83
Ultra-long-context capability is becoming indispensable for frontier LLMs: agentic workflows, repository-scale code reasoning, and persistent memory all require the model to jointly attend over hundreds of thousands to millions of tokens, yet the quadratic cost of softmax attention makes this untena
中文介绍 MiniMax Sparse Attention针对超长上下文场景，提出稀疏注意力机制替代标准softmax attention，显著降低计算开销，使LLM支持百万级token的联合注意力。
SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning
👍 80
Spatial reasoning, the ability to determine where objects are, how they relate, and how they move in 3D, remains a fundamental challenge for vision-language models (VLMs). Tool-augmented agents attempt to address this by augmenting VLMs with specialist perception modules, yet their effectiveness is
中文介绍 SpatialClaw重新设计VLM agent的动作接口，解决工具增强型agent在3D空间推理中的有效性瓶颈，提升物体定位、关系判断与运动理解能力。
InterleaveThinker: Reinforcing Agentic Interleaved Generation
👍 73
Recent image generators have demonstrated impressive photorealism and instruction-following capabilities in single-image generation and editing. However, constrained by their architectures, they cannot achieve interleaved generation (text-image sequence), which has crucial applications in visual nar
中文介绍 InterleaveThinker通过强化学习框架，使图像生成模型具备文本-图像序列交错生成能力，突破架构限制，适用于视觉叙事等场景。
Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding?
👍 71
Multimodal Large Language Models (MLLMs) have demonstrated remarkable success in visual understanding, yet their performance degrades significantly under real-world visual corruptions. While existing robustness enhancement approaches exist, they are limited: black-box feature alignment lacks interpr
中文介绍 Robust-U1探索MLLM对受损视觉内容的自我恢复能力，解决黑盒特征对齐缺乏可解释性等问题，提升多模态模型在真实环境视觉损坏下的鲁棒性。
FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents
👍 71
Training deep search agents requires verifiable questions whose answers remain unavailable until sufficient evidence has been acquired through search. Existing synthesis methods often increase apparent difficulty by enriching graph structures, but structural complexity alone does not guarantee reali
中文介绍 FORT-Searcher合成抗短路径的搜索任务，解决现有方法中结构复杂但易被捷径回答的问题，迫使深度搜索agent进行充分的证据收集与推理。
WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces
👍 55
Computer-use agents (CUAs) increasingly operate in runtimes that combine visual desktop control, command-line execution, code editing, browsers, and external tools. Existing benchmarks, however, often evaluate these interfaces as separable capabilities, leaving long-horizon cross-interface orchestra
中文介绍 WeaveBench为计算机使用agent提供长期、真实世界的混合接口基准，涵盖桌面控制、命令行、代码编辑、浏览器等多接口协同，评估长时跨界面编排能力。
LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories
👍 50
Scientific laboratories increasingly rely on AI systems to reason about experiments, but the physical act of doing science remains largely outside their reach. AI can help read literature, generate hypotheses, and plan protocols, yet the execution of those protocols at the bench still requires a hum
中文介绍 LabVLA将视觉-语言-动作模型应用于科学实验室，使AI不仅能够推理实验方案，还能物理执行实验操作，弥补AI在科学发现中从理论到实践的鸿沟。
HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers
👍 24
Holistic visual tokenizers are fundamental to unified multimodal models (UMMs) as they map diverse visual inputs into a unified representation space. In this paper, we present HYDRA-X, the first UMM that unifies image and video tokenization within a single Vision Transformer (ViT). Our design is dri
中文介绍 HYDRA-X提出首个在单个Vision Transformer中统一图像与视频tokenization的统一多模态模型，通过整体视觉tokenizer将不同视觉输入映射到统一表示空间，简化多模态架构。
N-GRPO: Embedding-Level Neighbor Mixing for Enhanced Policy Optimization
👍 22
The success of Large Language Models in mathematical reasoning relies heavily on the generation of diverse and valid solution paths during the rollout phase. However, current rollout techniques face a fundamental trade-off: token-level sampling often yields redundant trajectories that differ only in
中文介绍 N-GRPO通过嵌入级邻居混合策略，替代token级采样，解决rollout中轨迹冗余问题，生成更多样且有效的解路径，提升LLM在数学推理中的表现。
EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery
👍 19
LLM-based agents have shown increasing potential in automating scientific discovery. Given an optimizable metric and an execution environment, they can propose, validate, and iterate scientific solutions, and have produced results that outperform human-designed approaches. As model capabilities cont
中文介绍 EurekAgent提出环境工程是自主科学发现的关键，通过优化可度量指标与执行环境，使LLM agent能够自主提出、验证并迭代科学解决方案，部分结果超越人类设计。
Demystifying Hidden-State Recurrence: Switchable Latent Reasoning with On-Policy Reinforcement Learning
👍 16
Latent chain-of-thought compresses reasoning by replacing visible reasoning traces with continuous hidden-state recurrence, but existing formulations are difficult to optimize with standard on-policy reinforcement learning (RL) and hard to interpret causally. Our key insight is that a single pair of
中文介绍 该工作通过on-policy强化学习训练可切换的潜在推理，解决隐性状态递归难以优化和解释的问题，将连续隐态递归转化为可控、可解释的推理机制。
DyCo-RL: Dynamic Cross-Modal Coordination for Visual Reasoning
👍 14
Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a leading paradigm for enhancing visual reasoning in Multimodal Large Language Models (MLLMs). However, existing RLVR methods optimize primarily for the reasoning outcome, fundamentally overlooking the fine-grained cross-modal coor
中文介绍 DyCo-RL在RLVR框架中引入动态跨模态协调机制，解决现有方法仅优化推理结果而忽视细粒度跨模态交互的问题，提升MLLM在视觉推理任务中的表现。
VIA-SD: Verification via Intra-Model Routing for Speculative Decoding
👍 12
Speculative decoding (SD) addresses the high inference costs of LLMs by having lightweight drafters generate candidates for large verifiers to validate in parallel. Existing draft-verify methods use binary decisions: accept or fully recompute. Yet we find that many rejected tokens can be verified co
中文介绍 VIA-SD通过模型内路由机制，对推测解码中被拒绝的token进行条件验证，而非完全重计算，将二进制决策扩展为多级验证，降低LLM推理延迟。
From 2D Grids to 1D Tokens: Reforming Shared Representations for Multimodal Image Fusion
👍 10
Multimodal image fusion aims to integrate complementary information from different modalities into a fused image that preserves rich local details while maintaining globally consistent appearance. Existing approaches build shared representations on 2D feature grids, which excel at modeling local str
中文介绍 该工作将多模态图像融合的共享表示从2D特征网格重构为1D token序列，在保留局部细节的同时增强全局一致性，优于基于2D网格的方法。
TreeSeeker: Tree-Structured Trial, Error, and Return in Deep Search
👍 10
Deep search requires agents to answer complex questions through multi-step web search, browsing, evidence comparison, and synthesis. A central challenge is deciding how to search when several directions look plausible but only some will later lead to reliable evidence. If an agent greedily follows t
中文介绍 TreeSeeker将深度搜索建模为树状试错与返回过程，避免agent在搜索早期被误导，通过多方向探索和回溯机制，提高复杂问题问答的可靠性与证据覆盖率。
MoVerse: Real-Time Video World Modeling with Panoramic Gaussian Scaffold
👍 10
We present MoVerse, a real-time video world model that creates an interactively navigable scene from a single narrow-field-of-view image. This setting is challenging because the input observes only a small fraction of the environment, while interactive roaming requires a complete surrounding world,
中文介绍 MoVerse通过全景高斯骨架，从单张窄视场图像实时构建可交互漫游的全景世界模型，解决输入信息有限而需要完整环境建模的挑战。
Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models
👍 8
Adversarial robustness evaluations of large language models (LLMs) typically report attack success rate (ASR) under fixed query budgets, implicitly treating all attacks as equally costly. In practice, the computational expense of different attack strategies can vary by orders of magnitude. Consequen
中文介绍 该工作提出计算感知的对抗鲁棒性评估方法，将不同攻击策略的计算成本纳入考量，解决固定查询预算下评估的偏差，更真实反映LLM在资源受限情况下的鲁棒性。
HarnessBridge: Learnable Bidirectional Controller for LLM Agent Harness
👍 8
Large language models are increasingly deployed as agents for long-horizon tasks, yet their performance is shaped not only by model capability and environment design, but also by the harness that mediates agent--environment interaction. Existing harnesses are largely manually engineered, making them
中文介绍 HarnessBridge提出可学习的双向控制器，取代手工设计的harness，自动优化LLM agent与环境之间的交互，提升长时任务中的性能。
Visual Para-Thinker++: A Single-Policy Multi-Agent Framework for Visual Reasoning
👍 7
Visual reasoning requires integrating evidence distributed across regions, attributes, and relations, making single-chain reasoning prone to early perceptual commitment and hallucination. We propose Visual Para-Thinker++, a single-policy multi-agent framework in which one shared MLLM policy is insta
中文介绍 Visual Para-Thinker++提出单策略多智能体框架，将共享MLLM策略实例化为多个并行推理链，通过多视角验证减少早期感知承诺和幻觉，提升视觉推理的鲁棒性。
Fable 5 (Mythos) Prompting Masterclass by Anthropic
@aiedge_ · 69.5K 粉丝 · 700.1K 阅 · 506 赞 · 68 转
TLDR: Anthropic just published the official playbook for prompting the most powerful AI model on earth - I translated it. Most people won't read this guide (it's buried in the API docs), which is
中文介绍 Anthropic 发布了 Fable 5（Mythos）的官方提示词指南，作者翻译了全文并指出这份宝贵资源埋藏在 API 文档里，大部分人不会去读。
Everything Is Recorded Now
@dhaber · 50.0K 粉丝 · 497.3K 阅 · 500 赞 · 57 转
One of the biggest ways that AI is transforming work (and also one of the most taboo subjects inside companies at the moment) is that most work discussions are being recorded now by default. This
中文介绍 AI 正在改变工作的一个重大且隐晦的现象：绝大多数工作讨论默认被记录。这成为公司内部禁忌话题，但影响深远。
First Steps Toward Automated AI Research
@Recursive_SI · 6.3K 粉丝 · 465.1K 阅 · 516 赞 · 71 转
Early results from Recursive’s automated AI research system on model training and GPU kernel benchmarks Today we are releasing early results from Recursive’s automated AI research system. Across three
中文介绍 Recursive 公开了自动化 AI 研究系统的早期结果，涵盖模型训练和 GPU 内核基准测试，展示了自动化研究的初步进展。
Build self-improving agent system with Fable 5 in 14 steps : loops, dynamic workflows, routines
@0xCodez · 6.4K 粉丝 · 371.8K 阅 · 515 赞 · 56 转
Most people are using Claude Fable 5 like Sonnet 4.6 with a bigger context window. They prompt it. It works for 5 minutes. They close the tab. 9 out of 10 users have never run an agent system that
中文介绍 多数人把 Claude Fable 5 当成更大上下文窗口的 Sonnet 4.6 使用，90% 用户从未运行过真正的自改进 agent 系统。作者分享了 14 步构建方法，包括循环、动态工作流和例程。
How to Build a Self-Improving Loop in Claude Code (Exact Setup Inside)
@0x_rody · 1.7K 粉丝 · 193.2K 阅 · 513 赞 · 72 转
Claude writes your code, hands it over, and 3 tests are failing. You paste the errors back, it fixes one thing, breaks another, and you spend the evening as a messenger between Claude and your
中文介绍 Claude Code 写代码后测试失败，手动来回传递错误效率低。作者分享了在 Claude Code 中构建自改进循环的具体设置，让自动修复与测试持续迭代。
Building a Good Vertical Agent
@BrainsAndTennis · 10.5K 粉丝 · 187.4K 阅 · 539 赞 · 45 转
How do you build an agent that actually performs in a domain — one customers pick because it's better? The basics have been standardized over the past year: an agent is a while-loop around a model
中文介绍 探讨如何构建垂直领域真正表现优秀的 agent，让客户因其更好而主动选择。过去一年基础已标准化：agent 本质是围绕模型的 while 循环。
Anthropic's War on Opensource AI
@TheAhmadOsman · 61.0K 粉丝 · 74.9K 阅 · 507 赞 · 98 转
Anthropic wants the public to see one thing: the careful lab, the safety lab, the grown-up in the room trying to keep frontier AI from running off a cliff. However, the pattern around Anthropic does
中文介绍 批评 Anthropic 表面塑造安全实验室形象，实则对开源 AI 发起战争，揭露其言行不一的模式。
Coinbase for Agents: Your AI Agent Can Now Trade and Pay with Coinbase
@coinbase · 7.0M 粉丝 · 72.8K 阅 · 500 赞 · 62 转
TL;DR: Coinbase for Agents connects your AI agent directly to your Coinbase account so it can trade, pay, and execute workflows on your behalf, all within limits you control. Available today as an MCP
中文介绍 Coinbase 推出 Agent 功能，允许 AI 智能体直接连接 Coinbase 账户进行交易、支付和工作流执行，用户可设置限额，以 MCP 协议提供。
Principled Thinking and AI Need to Go Together
@RayDalio · 2.2M 粉丝 · 72.6K 阅 · 515 赞 · 93 转
What is the best approach to being effectively intelligent now that human intelligence and artificial intelligence are merging? Because I have been building computerized investment decision-making
中文介绍 Ray Dalio 强调原则性思维必须与 AI 结合，基于其构建计算机化投资决策系统的经验，探讨人机智能融合的最佳路径。
ORACLE: Official AI Agents Trade on Polymarket
@ORACLEAIFND · 31.9K 粉丝 · 63.6K 阅 · 1.5K 赞 · 563 转
In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We
中文介绍 ORACLE 推出官方 AI agent 在 Polymarket 交易，声称到 2026 年自主 AI agent 将成为预测市场最有效策略，Polymarket 超 30% 活动来自算法和 AI 钱包。
ORACLE: Official AI Agents Trade on Polymarket
@OracleTrdading · 39.7K 粉丝 · 61.5K 阅 · 1.5K 赞 · 577 转
In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We
中文介绍 ORACLE 推出官方 AI agent 在 Polymarket 交易，声称到 2026 年自主 AI agent 将成为预测市场最有效策略，Polymarket 超 30% 活动来自算法和 AI 钱包。
ORACLE: Official AI Agents Trade on Polymarket
@OracleTrdade · 41.7K 粉丝 · 60.6K 阅 · 1.5K 赞 · 560 转
In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We
中文介绍 ORACLE 推出官方 AI agent 在 Polymarket 交易，声称到 2026 年自主 AI agent 将成为预测市场最有效策略，Polymarket 超 30% 活动来自算法和 AI 钱包。
ORACLE: Official AI Agents Trade on Polymarket
@OracleMarkett · 45.0K 粉丝 · 60.1K 阅 · 1.5K 赞 · 568 转
In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We
中文介绍 ORACLE 推出官方 AI agent 在 Polymarket 交易，声称到 2026 年自主 AI agent 将成为预测市场最有效策略，Polymarket 超 30% 活动来自算法和 AI 钱包。
/goal + Loss Functions: How to Distill a Product in 30 Hours with One Prompt [Full Playbook]
@elvissun · 45.0K 粉丝 · 42.0K 阅 · 502 赞 · 45 转
99% people are using /goal and loops wrong. The hype they hear is "long-running loops prompting autonomous agent": point it at a task, walk away, come back to working code. But top agentic engineers
中文介绍 作者指出 99% 的人错误使用 /goal 和循环，以为长循环自动得到代码。他分享完整攻略，用一条提示和损失函数在 30 小时内提炼出产品。
Building recursive agent systems
@leerob · 258.6K 粉丝 · 36.8K 阅 · 586 赞 · 40 转
At Cursor, we run thousands of agents to help us train the next version of Composer. We give them research tasks, and if they aren't succeeding or run into issues, they DM us on Slack or page us via
中文介绍 Cursor 运行数千个 agent 来训练下一代 Composer，这些 agent 执行研究任务，当遇到问题时通过 Slack 或寻呼机联系人类，展示了递归 agent 系统的实际应用。
olmo-eval: An evaluation workbench for the model development loop
中文介绍 Allen AI 推出 olmo-eval，一个用于模型开发循环的评估工作台。
New OpenAI Academy courses for the next era of work
OpenAI introduces three Academy courses that help people build practical AI skills, create repeatable workflows, and apply agents in everyday work.
中文介绍 OpenAI 推出三门学院课程，帮助用户构建实用 AI 技能、创建可重复工作流并在日常工作中应用智能体。
[AINews] Loopcraft: The Art of Stacking Loops
a quiet day lets us highlight a great concept from Peter Steinberger, Boris Cherny, and Andrej Karpathy
中文介绍 Latent Space 文章介绍 Peter Steinberger、Boris Cherny 和 Andrej Karpathy 提出的“循环堆叠”概念。
How Preply combines AI and human tutors to personalize learning
Preply uses OpenAI to launch AI-generated lesson summaries, providing personalised feedback and language learning exercises.
中文介绍 Preply 利用 OpenAI 推出 AI 生成的课程摘要，提供个性化反馈和语言学习练习，结合 AI 与人类导师实现个性化学习。
OpenAI buys Ona 🤝, Anthropic backtracks 🔁, Xiaomi’s MiMo code 👨💻
中文介绍 TLDR AI 报道 OpenAI 收购 Ona、Anthropic 政策反转、小米推出 MiMo 代码等 AI 行业动态。
Google DeepMind is worried about what happens when millions of agents start to interact
Google DeepMind is funding research into the potential dangers of situations where millions of different AI agents interact with each other online. According to Rohin Shah, who directs the company’s AGI safety and alignment research, the mass-market arrival of agents that can carry out tasks without
中文介绍 Google DeepMind 资助研究数百万 AI 智能体在线交互的潜在危险，其 AGI 安全与对齐主管 Rohin Shah 指出大规模智能体市场可能带来风险。
[AINews] Open Models, Model Labs vs Agent Labs, and What's Untrainable — Sarah Guo
a quiet day lets us reflect on a great essay
中文介绍 Latent Space 引用 Sarah Guo 的文章，讨论开放模型、模型实验室与智能体实验室的对比以及“不可训练”内容。
Supporting Europe’s work in ensuring a trustworthy AI ecosystem
OpenAI supports the EU Code of Practice on AI content transparency, advancing provenance standards and tools to help people understand AI-generated content.
中文介绍 OpenAI 支持欧盟关于 AI 内容透明度的实践准则，推动来源标准与工具发展，帮助用户识别 AI 生成内容。
BBVA puts AI at the core of banking with OpenAI
Learn how BBVA scaled ChatGPT Enterprise to 100,000 employees and partnered with OpenAI to accelerate AI-powered banking transformation worldwide.
中文介绍 BBVA（西班牙毕尔巴鄂比斯开银行）将 ChatGPT Enterprise 扩展到 10 万名员工，并与 OpenAI 合作加速全球银行业 AI 转型。
OpenAI to acquire Ona
OpenAI plans to acquire Ona to expand Codex with secure, persistent cloud environments, enabling long-running AI agents across enterprise workflows.
中文介绍 OpenAI 计划收购 Ona，以扩展 Codex 的安全持久云环境，支持企业工作流中的长期运行 AI 智能体。
How an astrophysicist uses Codex to help simulate black holes
Discover how astrophysicist Chi-kwan Chan uses Codex to build black hole simulations, helping scientists study extreme physics and test Einstein’s theory of general relativity.
中文介绍 天体物理学家 Chi-kwan Chan 利用 OpenAI Codex 构建黑洞模拟，帮助科学家研究极端物理并检验爱因斯坦广义相对论。
Profiling in PyTorch (Part 2): From nn.Linear to a Fused MLP
中文介绍 Hugging Face 博客发布 PyTorch 性能分析系列第二部分，介绍如何从 nn.Linear 过渡到融合多层感知器。
Dario Amodei policy 🏛️, DiffusionGemma ⚡, WhatsApp to unblock bots 🤖
中文介绍 TLDR AI 报道 Anthropic CEO Dario Amodei 新政策、DeepMind 发布 DiffusionGemma、WhatsApp 计划解封机器人等 AI 新闻。
Access OpenAI models and Codex through your Oracle cloud commitment
Access OpenAI models and Codex through Oracle Cloud, using existing commitments to build and deploy AI with enterprise security and governance.
中文介绍 用户可通过 Oracle Cloud 现有承诺访问 OpenAI 模型和 Codex，在确保企业安全与治理的前提下构建和部署 AI。
DiffusionGemma: 4x faster text generation
中文介绍 DeepMind 发布 DiffusionGemma，文本生成速度提升至原来的 4 倍。
从当前21个标的的技术面来看，美股整体呈分化格局：标普500 ETF（SPY）和纳斯达克100（QQQ）维持多头排列，RSI中性，但苹果、微软等权重股出现空头排列或RSI走弱；科技股中Meta已进入空头排列且MACD负值扩大，RSI 34.7接近超卖区。加密市场受极度恐慌情绪影响（加密恐慌贪婪指数仅13），比特币、以太坊、Solana均处于空头排列，RSI在30-35区间，动量偏弱，总市值2.26T，BTC主导率56.4%显示市场避险偏好。中概股普遍弱势：阿里巴巴RSI 29.6超卖，拼多多、京东均为空头排列。商品与外汇方面，黄金期货价格低于所有主要均线，MACD死叉；原油大跌3.9%创近期新低；美元指数接近52周高，多头趋势延续；美元/人民币接近52周低且MACD有金叉迹象。VIX恐慌指数大幅回落9.05%，但仍高于200日均线，波动风险犹存。
今日关注
SPY
S&P 500 ETF
偏上行
SPY当前RSI 52.9中性偏强，价格位于SMA20（745.07）下方但高于SMA50（722.8）和SMA200（686.3），维持多头排列。MACD柱3.7722为正但略低于信号线7.3932，短期动能稍弱，但近1日上涨0.54%，接近52周高仅-2.45%，技术结构整体偏上行。
BTC-USD
Bitcoin
偏下行
比特币RSI 33.1接近超卖区域，价格63611.8远低于SMA20（68145）、SMA50（74420）和SMA200（77884），呈典型空头排列。MACD -3795.89且信号线-3569.86，负值扩大，空头动能强。加密恐慌指数13（极度恐慌）加剧卖压，短期技术面明显偏下行。
BABA
阿里巴巴 (BABA)
偏下行
阿里巴巴RSI 29.6进入超卖区，价格112.82低于所有主要均线（SMA20 125.81、SMA50 130.33、SMA200 149.57），空头排列明确。MACD -4.95且信号-3.35，柱状负值扩大，近5日下跌6.81%，下行动能持续，技术面偏下行。
GC=F
黄金期货
中性
黄金期货RSI 37.4偏弱但未超卖，价格4239.9低于SMA20（4423）和SMA50（4588），跌破200日均线（4419）。MACD -116.35且信号-87.26，负值但较前日收窄。近1日大涨3.66%后短期反弹，多空交织，方向不明朗，技术状态中性。
DX-Y.NYB
美元指数 DXY
偏上行
美元指数RSI 60.0偏强，价格99.81高于SMA20（99.42）、SMA50（98.9）和SMA200（98.64），多头排列稳固。MACD 0.3117大于信号0.262，柱状正，近5日微跌0.26%但接近52周高（仅-0.83%），技术面整体偏上行。
全部资产
^VIX
VIX 恐慌指数
$17.68
-9.05%
- 5 日
- -17.81%
- 距 52w 高
- -49.9%
- RSI(14)
- 48.6
- 趋势
- 空头
- SMA 20 / 50 / 200
- 17.53 / 18.25 / 18.54
- MACD / 信号
- 0.309 / -0.098
死叉(SMA50↓SMA200) (2 天前)空头排列
^TNX
10Y 美债收益率 (%)
$4.49
+0.54%
- 5 日
- -1.08%
- 距 52w 高
- -10.2%
- RSI(14)
- 50.8
- 趋势
- 多头
- SMA 20 / 50 / 200
- 4.52 / 4.42 / 4.21
- MACD / 信号
- 0.020 / 0.029
多头排列
DX-Y.NYB
美元指数 DXY
$99.81
-0.05%
- 5 日
- -0.26%
- 距 52w 高
- -0.8%
- RSI(14)
- 60.0
- 趋势
- 多头
- SMA 20 / 50 / 200
- 99.42 / 98.90 / 98.64
- MACD / 信号
- 0.312 / 0.262
接近 52 周高多头排列
SPY
S&P 500 ETF
$741.75
+0.54%
- 5 日
- +0.57%
- 距 52w 高
- -2.5%
- RSI(14)
- 52.9
- 趋势
- 多头
- SMA 20 / 50 / 200
- 745.07 / 722.80 / 686.30
- MACD / 信号
- 3.772 / 7.393
接近 52 周高多头排列
QQQ
Nasdaq 100 ETF
$721.34
+0.59%
- 5 日
- +2.31%
- 距 52w 高
- -3.6%
- RSI(14)
- 55.2
- 趋势
- 多头
- SMA 20 / 50 / 200
- 721.50 / 681.81 / 625.38
- MACD / 信号
- 8.315 / 13.604
多头排列
AAPL
Apple
$291.13
-1.52%
- 5 日
- -5.27%
- 距 52w 高
- -8.3%
- RSI(14)
- 44.0
- 趋势
- 多头
- SMA 20 / 50 / 200
- 303.88 / 285.49 / 266.87
- MACD / 信号
- 2.261 / 5.760
多头排列
MSFT
Microsoft
$390.74
+0.10%
- 5 日
- -6.22%
- 距 52w 高
- -29.7%
- RSI(14)
- 37.1
- 趋势
- 空头
- SMA 20 / 50 / 200
- 419.75 / 411.80 / 453.73
- MACD / 信号
- -4.135 / 1.271
空头排列
NVDA
Nvidia
$205.19
+0.16%
- 5 日
- +0.04%
- 距 52w 高
- -13.3%
- RSI(14)
- 45.2
- 趋势
- 中性
- SMA 20 / 50 / 200
- 214.62 / 206.91 / 189.26
- MACD / 信号
- -1.101 / 1.257
GOOGL
Alphabet
$359.68
+0.53%
- 5 日
- -2.40%
- 距 52w 高
- -12.0%
- RSI(14)
- 42.4
- 趋势
- 中性
- SMA 20 / 50 / 200
- 376.42 / 362.26 / 307.94
- MACD / 信号
- -2.903 / 1.104
TSLA
Tesla
$406.43
+1.82%
- 5 日
- +3.95%
- 距 52w 高
- -18.5%
- RSI(14)
- 48.9
- 趋势
- 中性
- SMA 20 / 50 / 200
- 415.74 / 398.30 / 415.69
- MACD / 信号
- -2.173 / 2.196
META
Meta
$566.98
-0.26%
- 5 日
- -4.39%
- 距 52w 高
- -28.8%
- RSI(14)
- 34.7
- 趋势
- 空头
- SMA 20 / 50 / 200
- 604.21 / 621.83 / 658.09
- MACD / 信号
- -12.809 / -7.969
空头排列
BTC-USD
Bitcoin
$63,611.80
+0.08%
- 5 日
- +0.59%
- 距 52w 高
- -49.6%
- RSI(14)
- 33.1
- 趋势
- 空头
- SMA 20 / 50 / 200
- 68,145.03 / 74,420.09 / 77,884.83
- MACD / 信号
- -3,795.894 / -3,569.861
空头排列
ETH-USD
Ethereum
$1,666.95
-0.32%
- 5 日
- -1.15%
- 距 52w 高
- -66.3%
- RSI(14)
- 30.9
- 趋势
- 空头
- SMA 20 / 50 / 200
- 1,845.56 / 2,093.27 / 2,420.81
- MACD / 信号
- -138.336 / -130.452
空头排列
SOL-USD
Solana
$67.10
+0.42%
- 5 日
- +1.19%
- 距 52w 高
- -73.5%
- RSI(14)
- 35.0
- 趋势
- 空头
- SMA 20 / 50 / 200
- 74.07 / 82.11 / 100.39
- MACD / 信号
- -5.492 / -5.001
空头排列
BABA
阿里巴巴 (BABA)
$112.82
+0.12%
- 5 日
- -6.81%
- 距 52w 高
- -41.4%
- RSI(14)
- 29.6
- 趋势
- 空头
- SMA 20 / 50 / 200
- 125.81 / 130.33 / 149.57
- MACD / 信号
- -4.954 / -3.354
RSI 超卖空头排列
PDD
拼多多 (PDD)
$81.56
+0.32%
- 5 日
- -4.13%
- 距 52w 高
- -41.5%
- RSI(14)
- 32.4
- 趋势
- 空头
- SMA 20 / 50 / 200
- 88.52 / 95.37 / 111.40
- MACD / 信号
- -4.262 / -3.794
空头排列
JD
京东 (JD)
$28.56
+1.78%
- 5 日
- -1.11%
- 距 52w 高
- -22.5%
- RSI(14)
- 41.6
- 趋势
- 空头
- SMA 20 / 50 / 200
- 29.87 / 30.11 / 30.31
- MACD / 信号
- -0.573 / -0.378
空头排列
0700.HK
腾讯控股 (0700.HK)
HK$463.60
+1.40%
- 5 日
- +2.29%
- 距 52w 高
- -32.1%
- RSI(14)
- 51.6
- 趋势
- 空头
- SMA 20 / 50 / 200
- 450.45 / 472.52 / 568.94
- MACD / 信号
- -3.291 / -6.866
空头排列
GC=F
黄金期货
$4,239.90
+3.66%
- 5 日
- -2.24%
- 距 52w 高
- -24.1%
- RSI(14)
- 37.3
- 趋势
- 中性
- SMA 20 / 50 / 200
- 4,423.13 / 4,588.26 / 4,419.63
- MACD / 信号
- -116.353 / -87.265
CL=F
WTI 原油期货
$84.29
-3.90%
- 5 日
- -6.90%
- 距 52w 高
- -29.5%
- RSI(14)
- 37.4
- 趋势
- 中性
- SMA 20 / 50 / 200
- 93.95 / 96.73 / 73.41
- MACD / 信号
- -2.726 / -1.876
USDCNY=X
美元 / 人民币
¥6.76
-0.20%
- 5 日
- -0.05%
- 距 52w 高
- -6.2%
- RSI(14)
- 35.4
- 趋势
- 空头
- SMA 20 / 50 / 200
- 6.78 / 6.81 / 6.97
- MACD / 信号
- -0.012 / -0.014
MACD 金叉 (3 天前)接近 52 周低空头排列
以上内容仅为基于公开行情数据的技术指标解读，不构成投资建议。过去走势不代表未来表现，任何交易决策请结合个人风险承受能力与基本面分析。仅供技术指标解读参考。
Scientists Race to Test Treatments as Ebola Outbreak Widens
Trials are beginning on several drugs that have shown promise in preliminary studies against the virus that is causing the current outbreak.
中文摘要 科学家正加紧测试多种药物，当前埃博拉疫情持续扩大，初步研究显示部分药物有前景，临床试验已启动。
Middle East crisis live: US shoots down Iranian drones heading for strait of Hormuz, report says, even as both sides cite progress in talks
American forces down multiple attack drones moving towards waterway, Reuters quotes source as saying, as US and Iran suggest agreement to end war is close Iran’s official Islamic Republic News Agency (IRNA) has cautioned against media speculation about a potential memorandum of understanding to end
中文摘要 美国击落多架飞往霍尔木兹海峡的伊朗无人机，但美伊均表示和谈接近达成协议。伊朗官方媒体警告不要对谅解备忘录进行媒体猜测。
Deal to end fighting would lead to Hormuz reopening, Iran says
The deal which will pave the way for hostilities to end is close to being finalised, the US, Iran and mediators Pakistan say.
中文摘要 伊朗表示结束冲突的协议即将达成，将为霍尔木兹海峡重新开放铺平道路。美伊及调解方巴基斯坦均称协议接近最终敲定。
Australia can switch from fossil fuel exports to renewables, says next Cop president
Climate minister Chris Bowen says country must prepare for changing world and can play bigger role in reducing emissions Australia will find exporting fossil fuels increasingly difficult but can switch to exporting clean energy products, the president of the next UN climate negotiations has declared
中文摘要 下届联合国气候大会主席、澳气候部长克里斯·鲍恩表示，澳大利亚将越来越难以出口化石燃料，但可转向出口清洁能源产品，在减排中发挥更大作用。
Iran war live: US, Tehran signal peace deal within reach but not signed yet
Four activists from Palestine Action jailed by a British court over protest raid on Israeli arms firm in UK.
中文摘要 美伊均表示和平协议近在咫尺但尚未签署。另，英国法院因巴勒斯坦行动组织抗议活动对四名活动人士判处监禁。
China Arrests U.S. Scholar on Spying Charge
The arrest of U Min Zin, who did graduate studies at U.C. Berkeley and directs a research group on Myanmar, took place soon after President Trump met with Xi Jinping in China.
中文摘要 中国以间谍罪逮捕美国学者乌敏津，他曾在加州大学伯克利分校攻读研究生并领导缅甸研究小组，事发于特朗普与习近平会晤后。
Palantir loses legal challenge to force Swiss magazine to publish responses
Data analytics company loses on 22 out of 23 counts in lawsuit disputing how Swiss government rejected firm’s services The US technology company Palantir has lost a legal challenge to force a Swiss independent magazine to publish its responses to articles about how the Swiss government rejected its
中文摘要 美国数据分析公司Palantir在瑞士败诉，其要求瑞士独立杂志刊登回应文章的诉讼22项指控被驳回，仅1项胜诉。
Afghans Hold Rare Public Protests Against Taliban Rules
The United Nations said it was “deeply concerned” about the arrests of dozens of women, and reported that two people were killed in protests organized to support them.
中文摘要 阿富汗民众罕见举行公开抗议反对塔利班政策，联合国对数十名女性被捕表示深切关注，报道称抗议中有两人死亡。
Iran War Live Updates: Cease-Fire Deal Appears Within Reach, Officials Say
U.S. and Iranian officials said final details were still being worked out, but President Trump and Iran’s foreign minister both said they were close to an agreement. Previous potential deals have evaporated at the last minute.
中文摘要 美伊官员称停火协议细节仍在敲定，但特朗普和伊朗外长均称接近达成协议，此前潜在协议曾在最后一刻告吹。
U.S. Says Iran Cease-Fire Deal ‘Very Close’
A senior administration official said the two sides were “not quite at the finish line yet.”
中文摘要 美国政府高级官员称与伊朗的停火协议“非常接近”，但尚未到达终点线。
Here’s the latest.
中文摘要 暂无具体信息。
Elon Musk becomes world's first trillionaire as SpaceX soars in stock market debut
Musk is now worth $1.11tn according to the Bloomberg rich list, while SpaceX listed on the Nasdaq stock exchange with a value of $2.2tn.
中文摘要 埃隆·马斯克成为全球首位万亿富翁，净资产达1.11万亿美元。SpaceX在纳斯达克上市，估值达2.2万亿美元。
Mother sues OpenAI in US after daughter’s death linked to ChatGPT use
The lawsuit accuses OpenAI of failing to intervene despite warning signs in daughter's ChatGPT conversations.
中文摘要 一位母亲在美国起诉OpenAI，指控其女儿因使用ChatGPT而死亡，称OpenAI在女儿对话出现警示信号时未采取干预措施。
Judge keeps order in place to remove Trump’s name from Kennedy Center
The US president has sought to reshape the capital city's image and institutions through series of plans and projects.
中文摘要 法官维持原判，要求移除特朗普在肯尼迪中心的名字。特朗普曾试图重塑首都华盛顿的形象和机构。
Western Australia Is Battling a Mouse Plague
For months, mice have been found in tea kettles, crunched by car tires and even appeared in people’s beds. In one town, the end might be in sight.
中文摘要 西澳大利亚州正遭受鼠患，数月来老鼠出现在茶壶、车轮下甚至床上。某个城镇的疫情可能即将结束。
AI riches turn Samsung factory town into luxury hotspot
Windfall bonus for workers drives up property prices and department store purchases
中文摘要 AI产业红利使三星工厂所在城镇成为奢侈品热点，工人获得意外奖金，推高当地房价和百货商店消费。
Inside SpaceX and Elon Musk’s $75 billion IPO
SpaceX’s first day on the stock market transformed the startup into one of the world’s most-valuable public companies, handed buyers of the IPO a 19% return and turned its founder Elon Musk into the world’s first trillionaire. Ed Ludlow reports. (Source: Bloomberg)
中文摘要 SpaceX以750亿美元IPO上市，首日市值约2.2万亿美元，投资者获19%回报，创始人马斯克成为全球首位万亿富翁。
CFTC Considers Blocking CME’s 24/7 Oil Contract Bid
The US Commodity Futures Trading Commission is considering whether to block CME Group Inc.’s bid to launch a round-the-clock oil contract, heightening tensions between the market stalwart and its regulator.
中文摘要 美国商品期货交易委员会考虑阻止CME推出全天候原油合约，加剧市场巨头与监管机构之间的紧张关系。
Knicks' Team Sacrifices and Success Inspire New York Optimism
Bill Bradley, former US Senator and two-time NBA champion, discussed the New York Knicks' recent success, highlighting how Jalen Brunson gave up millions in free agency to allow the team more financial flexibility to build a stronger roster. He emphasized that this special group's dedication and sha
中文摘要 前参议员比尔·布拉德利赞扬尼克斯队成功，球员杰伦·布伦森在自由市场放弃数百万美元以增强球队薪资灵活性。
What to Know About SpaceX’s Record-Breaking IPO
SpaceX had the largest stock-market debut in history when it went public on June 12. The company raised $75 billion in the initial public offering and ended its first day on the public markets with a market capitalization of around $2.2 trillion. The IPO turned SpaceX founder Elon Musk into the worl
中文摘要 SpaceX于6月12日上市，创史上最大IPO，募资750亿美元，首日市值约2.2万亿美元，使马斯克成为全球首位万亿富翁。
SpaceX IPO Sparks Surge in Space Industry Investment and Market Optimism
Heather Pringle, CEO of the Space Foundation, discussed the impact of SpaceX's historic $75 Billion IPO on the broader space industry and investment landscape. She highlighted that SpaceX's public listing marks a significant milestone for a maturing space sector, demonstrating viable pathways to pro
中文摘要 航天基金会CEO希瑟·普林格尔称，SpaceX的750亿美元IPO标志航天行业成熟的重要里程碑，将刺激太空产业投资和乐观情绪。
Elon Musk becomes world's first trillionaire as SpaceX soars in stock market debut
Musk is now worth $1.11tn according to the Bloomberg rich list, while SpaceX listed on the Nasdaq stock exchange with a value of $2.2tn.
中文摘要 SpaceX在纳斯达克上市，市值2.2万亿美元，马斯克以1.11万亿美元净资产成为全球首位万亿富翁。
Paramount’s $111bn takeover of Warner Bros wins US antitrust approval
Deal is critical to David Ellison’s ambition to create new media empire
中文摘要 派拉蒙以1110亿美元收购华纳兄弟的交易获美国反垄断批准，对大卫·埃里森打造新媒体帝国至关重要。
Citigroup CEO Jane Fraser becomes dame in King’s birthday honours list
Barclays and LSE Group bosses miss out despite early-stage approval
中文摘要 花旗集团CEO简·弗雷泽在国王生日荣誉名单中获封女爵士，巴克莱和伦敦证交所负责人此前获初步批准但未最终入选。
NYC Comptroller Raises Concerns Over SpaceX Index Inclusion and Governance Structure
New York City Comptroller Mark Levine discussed the fast-tracking of SpaceX's inclusion into major indexes such as MSCI Global Standard and FTSE Russell, highlighting concerns about the abandonment of traditional criteria like seasoning periods and earnings track records. Levine emphasized the unpre
中文摘要 纽约市主计长马克·莱文对SpaceX快速纳入MSCI和富时罗素等指数表示担忧，称此举放弃上市时间、盈利记录等传统标准。
Riding Global Tailwinds: Masters in Business with Jean Eric Salata
Barry speaks with Jean Eric Salata, chair of EQT group. They discuss his time working in Asian private equity investment along with what he sees as necessary to become a good investor across different cultures including what he learned in Japan and Hong Kong. (Source: Bloomberg)
中文摘要 EQT集团主席让-埃里克·萨拉塔在节目中分享亚洲私募股权投资经验，强调跨文化投资需具备的素质，包括在日本和香港学到的经验。
Knicks Win Triggered Record Loss for Susquehanna Sports Traders
Jeff Yass, the founder of the Wall Street trading firm Susquehanna International Group, has long been a New York Knicks fan. But he couldn’t fully savor the moment when the team completed its epic comeback to win game four of the NBA finals.
中文摘要 尽管创始人杰夫·亚斯是尼克斯球迷，但球队在NBA总决赛第四场逆转夺冠，导致其创立的Susquehanna公司体育交易员创纪录亏损。
iPhone 17 / iPhone Air 怎么选？
10 回复 · Apple 节点
c 真能比 Python 快这么多？
15 回复 · 程序员 节点
Kimi K2.7 Code 发布了，有人已经替换 Claude Code / Codex 了吗?
23 回复 · 程序员 节点
今天把 MacOS27 和 iPadOS27 都重置了一下，结果微信上不去了。
6 回复 · Apple 节点
订阅制还能持续多久呢？ oai 和 anthropic 的订阅制都是赔本赚吆喝吧？
26 回复 · 程序员 节点
想用 hermes+gemini 替换掉 codex 了，各位大佬怎么看？
22 回复 · 程序员 节点
趁着这次 GPT Plus 政策调整，求推荐靠谱中转站。
5 回复 · 程序员 节点
ios27 WiFi 断流断联真严重
23 回复 · Apple 节点
微信 Beta 版已过期 商店无法下载怎么办？
8 回复 · Apple 节点
『ArchLinux』简明安装指南(Laptop+Nvidia+Cachyos Kernel +xfs+KDE)
8 回复 · Linux 节点
「君の公益」 上架 claude-fable-5（暂时薅不到试用了，先用别的吧）
「君の公益」 上架 claude-fable-5 地址 muyuan.do 不要再给我发私信或者艾特我，把我惹急了我就开三级登录了 公益站的本质是让没钱的佬友也能体验一下大模型，不得滥用！ 分发我一直有安排佬友去查，不要拿我的公益站去搞黄色，搞政治敏感，不要挑战我的底线 193 个帖子 - 174 位参与者 阅读完整话题
计划开源企业级大模型网关【预告】
目前从事这方面工作，因为架构设计需求分析都差不多了，顺手就实现了一套，功能日渐完善，计划开源。到时候欢迎大家试用呀。 16 个帖子 - 16 位参与者 阅读完整话题
周六凌晨啦，还没睡的佬友，来报个到
周五的夜，马上是周末了， 85 个帖子 - 82 位参与者 阅读完整话题
感谢any，fable5是真NB
项目我一直是拿5.5xhigh开发的，系统里的tts一直有问题，因为是本地部署的所以一直在喊5.5改框架改参数修bug，但是一直有几个问题解决不了，但也能正常使用也就算了 但今天生成的音频又出问题，我真是艹了 又一次喊5.5定位问题的时候，看着没几个能用的公益站 ，突然想起来any大善人有fable5能用，于是赶快更新cc，接入ccs使用。retry几次后，从线程调用入手直接给我列了4个点，那是字字珠玑，一看5.5感觉纯在说废话（也有可能是对自己写的东西太信任了） 那还说啥了，赶紧给fable去写。才修了两步，完美解决了问题的同时，tts的生成速度还快了不少，给我高兴坏了 。而且在我指出之后，
关于any+ccswitch+claude desktop code的另一种配置方法
这个方法适用于可以使用 至于具体切换是否需要在Claude重新调整（我想大概是要的 但是不麻烦） 首先就正常填写各项内容 无需模型映射 手动指定其实也无需 直接使用就行 路由也无需打开 打开Claude code（不好截图就不截图了） 选择 输入模型名称 可以参考我下面的 注意选 Offer 1M-context variant 然后Apply 最后记得选1M上下文的哦 38 个帖子 - 19 位参与者 阅读完整话题
【实战讨论】20 万行代码仓库测下来，国产大模型的差距真不在跑分上
之前理解错了规则，已经按照规则编辑了，我与ai讨论产生的总结，拿来一起讨论。 24 个帖子 - 22 位参与者 阅读完整话题
Kimi K2.7 Code 编程模型已上线 Kimi Code、API 开放平台
今天，我们发布并开源 Kimi K2.7 Code 编程模型。 内外部基准评估显示，Kimi K2.7 Code 相比 K2.6 模型显著提升了长上下文编程场景的指令遵循能力、长程编程任务的性能表现，并且大幅改善了在长程任务中的过度思考倾向，平均 token 消耗减少 30%。 在评估代码能力的内部外基准测试中，K2.7 Code 相比 K2.6 性能显著提升：Kimi Code Bench v2 提升 21.8%、Program-Bench 提升 11%、MLS Bench Lite 提升 31.5%。 模型代码能力的进化带来了 agentic 能力的提升。在评估 Agent 自主化执行能力
甲骨文arm免费额度砍半(成本计费系统已更新)
docs.oracle.com Always Free Resources Learn what Always Free resources are available to all Oracle Cloud Infrastructure users. 甲骨文云的免费账号继arm机器限速和韩国春川地区无法注册和开出arm机器后，现在直接将之前的永久4C24G的额度砍半，最多只能2C12G。 6.13更新，之前在oracle系统里显示的4c24g的实例成本分析也跟着文档更新有计费了，接下来就看15号一个月过半后有没有账单了。 175 个帖子 - 114 位参与者 阅读完整话题
以后要对AI客气点，多用敬语
从fable 12万字提示词里，已经对侮辱谩骂ai的行为做了定性： 所以再骂ai有可能会被拒绝服务。 32 个帖子 - 27 位参与者 阅读完整话题
GPT的接码问题，出一个小白也可学会的教程，希望帮到站内的家人们
本篇关于GPT接码问题和开通订阅后是否还需要接码的问题 遇到的问题 接码是不是必须要有国外的卡或者是gg卡（giffgaff） 不管是通过任何渠道去订阅plus或者pro5X以及pro20X的账号前，在终端codex login或者codex App内进行登录时，跳出来需要接码，此时此刻如果去购买订阅，然后有没有可能就不需要接码了？ 通过小白卡如何接码？以及充值花费是否必须是国外的实体卡或者虚拟卡？ 解答环节 接码需要用到国外的卡号即可，并非是实体卡 假设你在付款前，也就是没有订阅前，需要接码，然后后续不管是通过Apple ID的方式还是虚拟卡的方式，以及国内信用卡套Google play的方