与「Zai」相关的搜索结果

A big Chinese lab reached out to me asking for help distilling Fable into their model. I said no. I told them to use the millions of books that have no copyright instead. Offered to help pick them and to help train the model a different way. They only wanted distillation, which tells you a lot about the mindset there. I've worked deep in this field and distillation is real. I wasn't convinced at one point. What I saw changed my mind completely. I'm the biggest supporter of open source. I'm for releasing weights and for the freedom to research and build, because humanity only grows when knowledge gets shared. But let's be logical. Nobody will convince me that labs shipping a model every month or less, each one a big leap, now can't ship a better one after three months. Kimi K3.1 Pro isn't a big leap over Kimi K3 and it's been about 100 days. DeepSeek still can't put out a respectable Pro model. Zai used to release their strongest model in under two months. It's been three with nothing. All I hear about Qwen 4 Max is that it's a letdown. Kimi K3.1 Pro and DeepSeek Flash both show traces of Astra distillation. The leak showed Moonshot used it, and so did DeepSeek. If this isn't the effect of Anthropic blocking distillation then what is it? Chinese labs need to get serious about using the talented researchers they have and give their models an identity. Right now their models at their best are a cheaper alternative. I want them out of that box. I want to see what makes them different. Anthropic has crossed lines too, no argument there. They destroyed millions of books, human knowledge that was in people's hands, and kept it for themselves. I'll never forgive them for that. But Chinese labs rely on distillation more than they rely on better data. I didn't work with them. Transparency and honesty matter to me, like you've known since I started here.

大肥鱼（DeepSeek Harness）真的牛逼啊！！这次直接把自媒体人的搞钱工作流做成了一个 AI 工作台！！ 一个插件，把自媒体团队的活儿全给串起来了 以前做自媒体，一个人恨不得长出六只手： 找选题、追热点、写文案、做封面、找配图、剪视频、排期、数据复盘…… 每天不是在创作，就是在各种软件之间反复横跳 DeepSeek Harness 里直接有了一个自媒体创作者工作台：dsh-media-desk 这几个功能真的有点东西： 1️⃣ 灵感收集：突然想到一个选题，一句话就能记录，随时沉淀创作素材 2️⃣ 脚本生产：从灵感到内容脚本，让 AI 帮你完成创作初稿 3️⃣ 视觉设计：封面、配图、视频分镜，内容包装一条龙 4️⃣ 内容资产库：历史作品统一归档，优秀内容可以持续复用 5️⃣ 发布排期：选题、创作、发布计划统一管理，不再靠脑子记 6️⃣ 数据复盘：把内容运营流程串起来，让每次创作都有积累 最让我兴奋的是它背后的思路： 不是让 AI 帮你写一篇帖子，而是让 AI 帮你搭建一套持续生产内容的流水线 以前一个人做自媒体，要同时扮演编辑、文案、设计师、视频策划和运营 现在一个人负责创意、判断和方向，Agent 辅助完成大量生产工作 一个人也能逐渐拥有一支 AI 内容团队 尤其是做 X、公众号、短视频、个人 IP 和商业内容的朋友，这种工作台值得研究 未来真正有竞争力的创作者，可能不是每天写得最多的人，而是最早建立起 AI 内容生产系统的人 Agent 时代，一个人就是一支内容团队，这才是真正的生产力升级！ GitHub 开源地址： 大肥鱼这波，确实把创意玩明白了！！ 安装使用教程看视频

🚨重磅！GLM 5.5最快将于本周发布！将掀起开源模型全新高度！ 智谱Zai创始人唐杰清华早已暗示两大飞轮RSI、后训练将铸就 GLM-5.5 成为无敌的存在！ 🔹飞轮已跑通：500M代码仓库→千万级Issue-PR合成→GLM-5 Agentic Engineering；合成沙盒→GLM-5.1自主执行；Agent自己造动态环境→GLM-5.2长程任务；网络安全环境，漏洞+PoC+Exploit→GLM-5.3。每一步都在用上一代模型生产下一代的训练数据。 🔹数据墙被击穿：Chinchilla定律明确，固定数据无法消除数据依赖损失。精选Web数据（RefinedWeb）远超原始语料，为飞轮提供高质量燃料。 🔹GLM 5.5是闭环产物：GLM-5.3已驱动Infra Agent在10万国产卡上自己优化推理系统，两周吞吐提升3倍+。飞轮继续转，5.5的后训练+自我进化环境只会更硬核，量化金融、代码沙盒、长程Agent全线拉满。 从Vibe Coding到Agentic Engineering，再到数据飞轮自我强化，5.5不是简单迭代，是这套系统第一次真正闭环后的产物。 #GLM55# #智谱# #唐杰# #数据飞轮# #RSI# #AgenticAI# #开源大模型# #清华# #AI# #Zai#

终于有人把这事做了！不用碰 Python，零依赖 JS/TS 直接拿 A股/港股/美股/基金/期货/期权全市场行情 stock-sdk，1.9K stars，内置 92 个 MCP 工具直接喂给 AI。 以前前端想做行情看板，行情工具基本全是 Python 生态，要么自己起后端，要么忍受财经接口 GBK/并发/批量各种坑。 stock-sdk 不一样，v2 架构大改，命名空间 API + 统一符号模型，零依赖，浏览器和 Node 18+ 双端直接跑。 核心能力： 1️⃣ 一套 API 覆盖 A 股、港股、美股、公募基金、期货、期权，实时行情、K 线、分钟线都能直接调 2️⃣ 资金流、龙虎榜、融资融券、大宗交易、涨跌停池、盘口异动、板块资金这些 A 股常用数据也做进去了 3️⃣ 自带 MACD、RSI、KDJ、BOLL、ATR、SAR 等技术指标，还能直接识别指标信号 4️⃣ 内置 Screener + 本地回测，不只是拿数据，筛股和策略验证也能直接做 5️⃣ 筹码分布也已经支持，A 股 / 港股 / 美股都能本地计算获利比例、平均成本和筹码区间 6️⃣ 最关键的是 MCP 已经原生做好了，Claude、Codex、Cursor 可以直接调用这些金融工具，不用你再包一层接口 我觉得它现在最有意思的一点，是已经从“股票数据包”慢慢变成了 Agent 的金融基础设施。 上面 AI 负责研究，下面 stock-sdk 负责行情、指标、选股和回测。 做股票 Agent、AI 投研，或者想给自己的量化系统补一套统一数据层，这个很值得留着。

🚨Ox Alpha has officially been revealed Zai have confirmed Ox Alpha is theirs and a new GLM model, with the weights dropping tonight Looks like the GLM 5.3 Flash theory aged pretty well

GLM-5.3 发布了 这周又是极度热闹的一周 先是 Grok 4.6，再是 DeepSeek V4 Pro 和 DeepSeek Harness (以及 API 涨价) 现在临近周末 @Zai_org 又带来了 GLM-5.3 我甚至都没提 Gemini 3.7 Flash 😂

From Mecha Hitler to SOTA rare-disease diagnosis in children? @SpaceXAI's @grok 4.6 has taken the 👑 on RareBench, edging out @AnthropicAI Claude Opus 5 for about 1/3 the cost. This was not on my 2026 bingo card! @deepseek_ai's new v4-pro-0813 model underperformed my expectations, v4-flash, and seemingly the entire internet's. We accessed using DeepSeek's 1P API on the day of release and I almost wonder if they didn't switch over their model endpoint correctly. We will re-benchmark and report back. @Zai_org has attracted a following with GLM5.2, but they, too, underperformed. This doesn't surprise me because when I compared GLM and @Kimi_Moonshot K3 for coding use-cases, I found Kimi substantially stronger, but the internet seems to love this model.

🔥 用户专享福利：GLM-5.2 明日开启限时 6 折优惠！ 旗舰性能，极致性价比！作为 @Zai_org 新一代旗舰模型与开源权重代表作，GLM-5.2 即将在 开启限时专属 6 折优惠！无缝覆盖 API 接入与 Web Chat 体验，帮你把顶尖大模型算力成本直接打下来！ 💡 为什么选择在 调用 GLM-5.2？ 🔹 全场景无缝接入： 官方API 与 Web 端同步享折扣，原厂API直连，极速响应，稳定可靠 🔹 极致成本优化： 输入低至 0.84，缓存读取仅需 0.168，大幅降低生产环境部署成本 🔹 旗舰全能表现： 在逻辑推理、复杂代码编写、长文本分析及 Agent 智能体调度上展现极强实力 ⏳ 锁定明天 8 月 12 日，来 感受顶级性能与超高性价比的完美结合！🎉