Nathan Lambert@natolambert · 4天前59

I quickly became friends with Arcee's leadership and can't help but root for their humble approach to building the open ecosystem. No nonsense licenses, no projecting, just enabling broad access to efficient intelligence. I'm happily supporting their research as an advisor.

译知名 AI 研究员 Nathan Lambert 宣布加入 Arcee 担任研究顾问。他称赞 Arcee 领导层以谦逊方式构建开放生态，提供无繁琐许可证、广泛可及的高效智能。Arcee 表示其加入恰逢开放模型生态面临日益压力之际，将为美国开放模型运动注入关键力量。

elvis@omarsar0 · 4天前67

This is just awesomeness from @cohere, @nickfrosst, and team. I so badly want a coding agent that just runs on my local machine. We are not too far now! Excited to get this to work with my @dair_ai coding agent in the next couple of days.

译Cohere发布了其首个开源编程模型North Mini Code。该模型小巧高效，专为智能体性能设计，并欢迎社区反馈。Elvis Saravia对此赞叹不已，期待尽快将其与自己的DAIR.AI编程智能体配合使用。

Yuchen Jin@Yuchenj_UW · 4天前47

While amazed by the Mythos/Fable benchmarks, I’m grateful Anthropic isn’t the only frontier AI company in the world. One company being able to build AI God is far riskier than multiple companies having that capability. Above all, I hope open-source AI keeps thriving.

译虽对Mythos/Fable基准测试感到惊叹，但我很感激Anthropic并非世界上唯一的前沿AI公司。 单一公司能构建AI之神，其风险远大于多家公司拥有这种能力。 最重要的是，我希望开源AI能持续繁荣。

Rohan Paul@rohanpaul_ai · 4天前64

Apodex-1.0-H just dropped a heavy-duty agent team for deep research Claims SOTA results by splitting web research across many agents and auditing every evidence chain before writing the answer. Treats deep research as a distributed systems problem for AI agents. Apodex uses an async agent team: an orchestrator assigns sub-agents separate contexts and tools, then fact-checker, conflict-reviewer, and draft-reviewer agents test weak claims. The real big deal is that Apodex is showing a possible “inference-time scaling” path for AI research, where better answers come not from one bigger model, but from many coordinated search agents, persistent traces, and a separate verification layer that audits the evidence before the final response is allowed to exist.

译Apodex-1.0-H 发布一个异步智能体团队，用于深度研究。协调者将子智能体分配到独立上下文和工具，再通过事实核查、冲突审查和草稿审查智能体检验弱主张。该方案将深度研究视为分布式系统问题，展示了推理时缩放路径：通过多个协调搜索智能体、持久追踪和独立验证层提升答案质量，而非依赖单一更大模型，并声称取得 SOTA 结果。

Claude@claudeai · 4天前47

Michael Truell (@mntruell) fell in love with coding at 12. The company he co-founded, @cursor_ai, went from 15 people to 700 in two years. Today, over 60% of the Fortune 500 build with its AI coding platform.

译Michael Truell（@mntruell）12 岁时爱上了编程。他联合创立的公司 @cursor_ai 在两年内从 15 人发展到 700 人。如今，超过 60% 的财富 500 强企业使用其 AI 编码平台。

Chubby♨️@kimmonismus · 4天前67

The Elder Scrolls 5 Fablewind (v2). Working inventory + items, working quests, level-system, mana/stamina/health working, buy/sell and a bit updated graphics. Lets see how good this geeds

译The Elder Scrolls 5 Fablewind (v2)。工作库存+物品，工作任务，等级系统，法力/耐力/生命值工作，买卖和一点更新的图形。 让我们看看这个有多好

Replit ⠕@Replit · 4天前65

Most people run a security scan for malicious packages before publishing a project But the risk starts the moment they're installed Today we're launching Package Firewall, built in partnership with Socket It blocks malware before it ever reaches your app

译大多数人在发布项目前会运行安全扫描以检测恶意包 但风险从安装的那一刻就已开始 今天，我们正式推出 Package Firewall，与 Socket 合作构建 它在恶意软件到达你的应用之前就将其拦截

🚨 AI News | TestingCatalog@testingcatalog · 4天前62

Inworld has cut prices across its Realtime Inference and Speech-to-Text services, repricing the open models it serves so consumer voice apps can run at scale more cheaply and for longer. With Realtime Inference, Speech-to-Text with voice profiling, and a Realtime API that now runs Gemma 4, DeepSeek, and MiniMax at around half the public rate behind a single OpenAI-compatible endpoint.

译Inworld 大幅降低实时推理、带语音特征分析的语音转文本（STT）以及 TTS 服务的 API 价格，将 Gemma 4、DeepSeek、MiniMax 等开源模型

🚨 AI News | TestingCatalog@testingcatalog · 4天前77

Google released DiffusionGemma, a new open model with up to 4x faster output! > Instead of predicting word-by-word, it generates entire blocks of text simultaneously. This lets the model self-correct and format complex markdown in real time. Same performance as Gemma 4 is a big deal. Wondering if it is based on the model previewed last year.

译Google发布了实验性开源模型DiffusionGemma，采用Apache 2.0许可。该模型不再逐token预测，而是同时生成整块文本，速度提升高达4倍。它能在生成过程中自我纠正，并实时格式化复杂Markdown。性能与Gemma 4相当。

elvis@omarsar0 · 4天前59

Every CEO and every engineer should be strategizing carefully to avoid AI model lock-in. More here: https://x.com/omarsar0/status/2063750411576844684?s=20

译Elvis Saravia指出每位CEO和工程师都应谨慎制定策略，避免AI模型锁定。引用推文以Anthropic为例，其做法是构建强大产品后单方面根据自身标准决定用户能否使用某些提示，此举虽合法但给企业带来业务连续性风险，尤其当各编码模型性能趋同时。建议转向开源且更便宜的替代方案以保持控制，并强调“控制层（control plane）”是自然解决方案。

Ethan Mollick@emollick · 4天前45

More true than ever.

译如果你在考虑接受一份工作，你可能想问你的token预算会是多少。这比以往更真实了。

Ethan Mollick@emollick · 4天前62

We need more real time data on how AI may be impacting the economy - this is a really useful addition.

译我们需要更多关于AI如何影响经济的实时数据——这是一个非常有用的补充。

宝玉@dotey · 4天前62

Claude Design 有一个完整的 Agent 运行环境，如果把它跟 Claude Code 比较的话，只是它基于浏览器和虚拟机来构建的。它有45个工具，有24个内置skills。运行方式也是基于 Agentic Loop。 所以它就是一个完整的 Agent Harness。

译Claude Design 具备完整 Agent 运行环境，基于浏览器和虚拟机，内置 45 个工具和 24 个 skills，采用 Agentic Loop。有团队质疑其仅为单个 Agent，宝玉回应：它就是一个完整的 Agent Harness。

Chubby♨️@kimmonismus · 4天前43

More realistic example of a one shotted game. Asked Fable 5 to recreate a game in the style of The Elder Scrolls 5 Morrowind. It one shotted quests, currencys and fighting, journal and minimap. And it worked.

译一个更现实的单次生成游戏的例子。要求Fable 5以《上古卷轴5：晨风》的风格重新创建一个游戏。它一次性生成了任务、货币、战斗、日志和小地图。而且它成功了。

Google AI Developers@googleaidevs · 4天前67

DiffusionGemma, our experimental open model released under an Apache 2.0 license, explores text diffusion, an exceptionally fast approach to text generation. Here’s how DiffusionGemma accelerates development: + Faster token output: By shifting the bottleneck from memory bandwidth to raw compute, the model generates up to 4x faster token output on dedicated GPUs + Accessible hardware footprint: Activates just 3.8B parameters during inference, fitting comfortably within 24GB-VRAM high-end consumer GPUs when quantized + Novel workflows: Parallel token generation enables self-correction, making it ideal for code infilling, in-line editing, and non-linear structures DiffusionGemma prioritizes speed over raw quality and accelerates best on compute-bound hardware (like @NVIDIAAI GPUs). Standard @GoogleGemma 4 remains recommended for production quality and memory-bound devices.

译Google AI 发布实验性开源模型 DiffusionGemma，采用 Apache 2.0 许可证。该模型基于文本扩散方法，将生成瓶颈从内存带宽转向计算，在专用 GPU 上 token 输出速度最高提升 4 倍。推理时仅激活 3.8B 参数，量化后可适配 24GB VRAM 消费级 GPU。并行 token 生成支持自我纠错，适用于代码填充、行内编辑等非线性结构。DiffusionGemma 优先速度而非极致质量，生产场景仍推荐标准 Gemma 4。

fofr@fofrAI · 4天前69

DiffusionGemma, where the LLM picks words all at once. Which is 4x faster. You can get started with the weights and instructions here: https://huggingface.co/google/diffusiongemma-26B-A4B-it

译DiffusionGemma，大语言模型一次性选出所有词。速度快4倍。 你可以从这里获取权重和说明开始使用： https://huggingface.co/google/diffusiongemma-26B-A4B-it

Emad@EMostaque · 4天前17

anyone hiring witches to save training runs

译有人在雇女巫来拯救训练运行吗？

AYi@AYi_AInotes · 4天前43

Damn，AGI到来之后，品味和审美一定是一个人最核心的竞争力和最强的护城河， Claude Fable 5遇到审美和品味高的，输出质量高到爆炸，这效果真的绝了😭

译天哪，AGI到来之后，品味和审美一定是一个人最核心的竞争力和最强的护城河， Claude Fable 5遇到审美和品味高的，输出质量高到爆炸，这效果真的绝了😭

🚨 AI News | TestingCatalog@testingcatalog · 4天前63

NoimosAI launched an all-in-one autonomous AI marketing team that runs strategy, content, and analysis end-to-end. > It connects to a brand's own apps and websites, blends that with external market data, and acts at the right time on its own. > Outputs land in a Feed for approval and can be routed to Slack, email, or Discord.

译NoimosAI 推出全自主 AI 营销团队，可端到端运行策略、内容创作与分析。它连接品牌自有应用和网站，融合外部市场数据，自主判断执行时机。所有产出汇集到 Feed 供审批，并支持路由至 Slack、Email 或 Discord。功能覆盖 SEO、社交媒体、外联、GEO 等，7×24 小时运行，无需人工干预。

François Chollet@fchollet · 4天前72

Some considerations that many folks seem not to get: 1. It can be a bubble even if the tech works. (For instance, if the tech doesn't have a high-demand use case.) 2. It can be a bubble even if the tech works and has strong product-market fit. (For instance, if the tech cannot be economically viable.) 3. It can be a bubble even if the tech works, has strong product-market fit, and has a path to eventual economic viability. (For instance, if profitability takes too long to achieve or makes margin/competition assumptions that fail to materialize.) 4. It can be a bubble even if the tech works, has strong product-market fit, and is currently highly profitable. (For instance, if demand has a hard ceiling and growth stops once the ceiling is reached.) 5. It can be a bubble even if the tech works, has strong product-market fit, is currently highly profitable, and has unlimited future demand. Literally all it takes for something to be a bubble is for lots of people to over-enthusiastically bet their money on it, and subsequently get panicky. Importantly, bubbles can be attached both to things that are completely hogwash, like the Metaverse, and to world-changing developments like the Internet or railways. Bubbles don't care. They're brought into existence by the thoughts and feelings of investors, not by actual tech or products. "The bubble has burst" doesn't mean "the tech didn't work" or "people stopped using the tech." It only means that people got panicky, investor money dried up, and valuations collapsed. Internet adoption didn't stop in 2000.

译Francois Chollet 指出，许多人误以为只要技术有效、有市场、能盈利甚至需求无限，就不可能是泡沫。实际上，泡沫的定义是投资者过度乐观地押注后恐慌，与技术的实际价值无关。无论技术是像元宇宙那样的噱头，还是像互联网或铁路这样的变革性发明，都可能产生泡沫。泡沫破裂并不意味着技术失败或用户停止使用，只是投资资金枯竭、估值崩溃——就像2000年互联网泡沫并未阻止互联网普及。

elvis@omarsar0 · 4天前42

So the message I am getting is that I can't use Fable to further accelerate AI research and education. No company will decide that for me. Just an absolutely sad day for the research community. As a developer, I already felt ignored, and now this. Damn it!

译所以我得到的信息是，我无法使用Fable来进一步加速AI研究和教育。 没有任何公司能替我做这个决定。 对于研究社区来说，这绝对是个令人悲伤的日子。 作为一个开发者，我已经感到被忽视，现在又这样。 该死！

elvis@omarsar0 · 4天前71

This is awesome! I am spending a lot of time on diffusion LLMs these days, so this is perfect timing. I feel like there are so many underexplored research questions around text diffusion. Weight available in HF.

译太棒了！我最近花了很多时间在研究扩散大语言模型上，所以这个时机恰到好处。我觉得文本扩散领域还有很多未被充分探索的研究问题。权重已在 HuggingFace 上可用。

AK@_akhaliq · 4天前46

ABot-Earth 0.5 Generative 3D Earth Model

译ABot-Earth 0.5 生成式3D地球模型

Rohan Paul@rohanpaul_ai · 4天前56

NoimosAI just launched an autonomous AI marketing team. It connects to a product’s apps, website, and data sources, then plans and runs growth work across SEO, GEO, social, outreach, and conversion. The product is built around a clear bottleneck: AI has made product building faster, but distribution still depends on traffic, leads, content, timing, and repeated testing. So NoimosAI is trying turn that marketing grind into a daily operating loop, where the system studies your business, suggests high-impact work, runs tasks 24/7, and sends finished outputs for approval.

译NoimosAI 推出全合一自主 AI 营销团队。连接产品应用、网站及数据源后，系统自动规划执行 SEO、GEO、社交媒体、外展和转化等增长工作，24/7 运行并将输出结果送交审批。产品基于一个关键洞察：AI 虽加速了产品构建，但分发仍依赖流量、线索、内容和反复测试，因此 NoimosAI 试图将营销流程转化为日常自动运营闭环。

Sundar Pichai@sundarpichai · 4天前75

DiffusionGemma is an open, experimental model that brings our text diffusion research to Gemma 4. It’s a racehorse 🏇achieving up to 4x faster inference by generating entire blocks of text simultaneously vs predicting token-by-token (word-by-word) output!

译DiffusionGemma 是一个开放的实验性模型，它将我们的文本扩散研究引入 Gemma 4。它是一匹赛马 🏇，通过同时生成整块文本（而非逐 token（逐词）预测输出）实现高达 4 倍更快的推理速度！

Epoch AI@EpochAIResearch · 4天前48

How should we think through various proposals for sharing the gains of AGI? According to @pawtrammell and @ansonwhho, the leading proposals for universal redistribution after AGI differ along a primary axis: how much direct control over capital they propose giving citizens. 🧵

译我们应该如何思考各种关于分享AGI收益的提案？ 根据@pawtrammell和@ansonwhho的观点，AGI后普遍再分配的主要提案在一个主要轴线上存在差异：它们提议给予公民多少对资本的直接控制权。🧵

Claude@claudeai · 4天前72

New from Code with Claude Tokyo: scheduled deployments and environment variables in vaults are in public beta in Claude Managed Agents, and dynamic workflows in Claude Code are generally available. Agents now run on a schedule, use your tools securely, and take on bigger jobs.

译Code with Claude Tokyo 新功能：定时部署和保险库中的环境变量已在 Claude Managed Agents 中进入公测，Claude Code 的动态工作流已正式可用。 Agent 现在可以按计划运行，安全使用你的工具，并承担更大的任务。

向阳乔木@vista8 · 4天前32

对比 Fable 5，忽然觉得把大模型变便宜这件事儿真的很重要。 真的要夸下DeepSeek，实在是便宜。 每月用 v4 flash 翻译不少东西，一看账单才花几十块。 把自己的 AI 雷达 Skill 变成网站，支持双语对照翻译和 AI 侧边栏对话。 下周开源，本周先上网站，输入DeepSeek API能用 AI 功能，也能人工点评

译作者对比 Fable 5 后强调，把大模型变便宜至关重要。他称赞 DeepSeek 极低的成本——每月使用 v4 flash 翻译大量内容，账单仅几十元。作者将自己的 AI 雷达 Skill 做成网站，支持双语对照翻译和 AI 侧边栏对话。网站本周先上线（输入 DeepSeek API 即可使用 AI 功能，也可人工点评），下周将开源。

Fei-Fei Li@drfeifei · 4天前20

Scientific research is fundamental to advancing civilization and helping people globally to solve the most critical problems, from medicine to materials, from brain science to physics, and much beyond. This is only possible when scientists have access to the best tools of the time to conduct scientific research, including having access to AI-based tools.

译科学研究是推动文明进步、帮助全球人民解决最关键问题的基础——从医学到材料，从脑科学到物理学，乃至更广泛的领域。这只有在对科学家拥有时代最佳工具（包括基于AI的工具）来进行科学研究时才有可能实现。

Chubby♨️@kimmonismus · 4天前61

Everyone is talking about basic chat apps, but desktop AI agents are completely shifting how we work in 2026. I’ve been testing Tencent WorkBuddy now that it’s finally going global. It dropped in China back in March and instantly hit number one for daily active PC users according to Tencent’s Q1 earnings. @TencentAl_News

译人人都在谈论基础聊天应用，但桌面AI智能体正在完全改变2026年的工作方式。 我已经测试了腾讯工作助手，它终于走向全球。它于3月在中国发布，根据腾讯Q1财报，其PC日活跃用户数立即排名第一。@TencentAl_News

Chubby♨️@kimmonismus · 4天前61

Most "AI marketing" tools are basically Zapier with a chat box. NoimosAI plays in a different league. It plans the strategy, creates the content, ships it across every channel and analyzes what worked, then adjusts on its own. You connect your stack and it starts working in your brand voice the same day. There's a 7-day free trial, so you can watch it run before you commit.

译Kim 介绍 NoimosAI 是不同于普通 Zapier+聊天框的 AI 营销工具：它能自动制定策略、创建内容、跨渠道发布、分析效果并自主调整。用户连接现有工具后当天即可按品牌语调工作，提供 7 天免费试用。引用推文称其为全球首个全自主 AI 营销团队，覆盖 SEO、社交、外展、GEO 等，24/7 无人值守运行。

Google DeepMind@GoogleDeepMind · 4天前72

DiffusionGemma is our new experimental open model with up to 4x faster output on dedicated GPUs. Instead of predicting word-by-word, it generates entire blocks of text simultaneously. This lets the model self-correct and format complex markdown in real time.

译DiffusionGemma 是我们新的实验性开放模型，在专用 GPU 上输出速度最高可提升 4 倍。 它不是逐词预测，而是同时生成整块文本。这让模型能够自我纠正，并实时格式化复杂 Markdown。

🚨 AI News | TestingCatalog@testingcatalog · 4天前40

Meta is working on support for Custom Instructions for Meta AI on the web.

译Meta 正在为网络版 Meta AI 开发自定义指令支持。

Kling AI@Kling_ai · 4天前25

What’s inside this football? ⚽

译这个足球里面有什么？⚽

SiliconFlow@SiliconFlowAI · 4天前58

If you need one model for agents, long context, and multimodal inputs — this is it. Meet @GoogleDeepMind 's Gemma 4 12B on SiliconFlow 🔥 💰Input / Output: $0.1 / $0.3 per 1M tokens on SiliconFlow 🛠️ 262K Context | Built-in Thinking | Native Tool Calling | 140+ Languages ✨ Encoder-free architecture: vision and audio inputs flow directly into the LLM backbone, reducing process latency 🧠 12B Size, 26B Brain: nearing Google's 26B performance, excel at multi-step reasoning and agentic workflows Try it on SiliconFlow ⬇️

译Google DeepMind 的 Gemma 4 12B 已在硅基流动上线，定价输入 $0.1/1M tokens，输出 $0.3/1M tokens。支持 262K 上下文、内置思考、原生工具调用及 140+ 种语言。采用无编码器架构，视觉和音频输入直接注入 LLM 主干，降低处理延迟。12B 参数但配备 26B “大脑”，性能接近 Google 26B 级别，擅长多步推理与智能体工作流。

Chubby♨️@kimmonismus · 4天前58

This is one of the most important parts of the Mythos 5 announcement: Drug design. Anthropic says its internal protein design experts used Mythos 5 to accelerate parts of the drug design process by around 10x. In one example, Mythos 5 used protein design and bioinformatics tools without (!) human assistance and matched or beat skilled human operators. It means the model could run parts of the early drug discovery loop itself: choosing binding sites, selecting tools, running protein design workflows, recovering from failures, and generating promising candidates. And here is what people still dont understand outside of our community: AI is moving from "assistant that explains science" to "agent that can actually execute parts of scientific work." The future of drug discovery may look less like one scientist manually testing ideas one by one, and more like thousands of AI-driven research loops running in parallel, with humans validating, interpreting, and deciding what actually moves forward. I keep thinking about what Demis Hassabis said: we are entering the golden age of science. There is probably no doubt about that anymore. Medicines are being developed faster and more precisely, and previously incurable diseases are becoming treatable.

译Anthropic内部蛋白质设计专家使用Mythos 5将药物设计部分流程加速约10倍。在一例中，Mythos 5无需人类协助即可自主使用蛋白质设计和生物信息学工具，匹配或超越熟练人类操作，能独立完成早期药物发现中的结合位点选择、工具调用、工作流运行、故障恢复和候选生成。在Gemini 3.5 Pro和GPT-5.6临近发布的背景下，Anthropic凭借Mythos/Fable进入新层级，给竞争对手带来压力，其IPO前景也因此提升。

Chubby♨️@kimmonismus · 4天前73

OpenAI appears to be preparing for a possible IPO within the next year, but Sam Altman is keeping the door open. If recursive self-improvement starts looking real, staying private could become the smarter move. At the same time, OpenAI’s enormous compute needs may push it toward public markets sooner, while the company is also preparing a new model, codenamed 5.6, described internally as a meaningful improvement over GPT-5.5. via The Information

译The Information 报道，OpenAI 正筹备未来一年内 IPO，CEO Sam Altman 持开放态度：若递归自我改进趋势显现，私有可能更明智；巨大计算需求也可能推动更早上市。公司同时准备代号 5.6 的新模型，内部描述为对 GPT-5.5 的有意义改进。本周早些时候，Altman 与首席科学家 Jakub Pachocki 已向员工发送备忘录讨论上述事宜。

OpenRouter@OpenRouter · 4天前66

Today we're launching the new Activity explorer on OpenRouter. It's the best way to see how much and your team are spending on every model, along with tokens, cache hit rate, agents, & trends. All updated in real time. See how our team is using Fable and other models 👇

译今天，我们在 OpenRouter 上推出了新的 Activity explorer。 这是查看你和团队在每个模型上花费了多少的最佳方式，还包括 token、缓存命中率、智能体以及趋势。所有数据实时更新。 看看我们的团队如何使用 Fable 和其他模型 👇

ChatGPT@ChatGPTapp · 4天前69

Go #MessiMode Upload a photo of yourself and try this prompt: “Make my hair the colors of my country flag but keep it natural-looking. If no country or image is provided, ask."

译Go #MessiMode 上传一张你的照片并尝试这个提示词：“将我的头发变成本国国旗的颜色，但要看起来自然。如果没有提供国家或图片，请询问。”

AYi@AYi_AInotes · 4天前71

怎么让Claude Fable 5做一个关于他自己的网站？ 这个提示词结构很顶，完全可以拿去让Fable做一个个人网站啊，效果真的很炸🤯 Prompt: 生成一个关于你自己的登录页面：Claude Fable 5， 你完全自由，但要确保它视觉吸引人，简洁，互动性强，带有动态动态效果，彩蛋， 搜索并使用2026设计趋势。

译推文分享了一个让 Claude Fable 5 生成关于自身登录页面的提示词结构。提示词要求完全自由，视觉吸引人、简洁、互动性强、包含动态效果和彩蛋，并搜索使用 2026 设计趋势。引用推文显示结果令人惊艳。作者认为该提示词结构同样适用于创建个人网站。