Link

一个非程序员的人机共生记录。
研究 AI，实践 AI，也观察 AI 正在怎样改变我们。
AI 丨 Crypto 丨资产配置

[Link @Link_gogo]

https://t.co/wHUq3ZDTse

[Link @Link_gogo]

@RlxemVHu5e34474 1

[Link @Link_gogo]

@Xaviertcrypto me

[Link @Link_gogo]

https://t.co/BuL9LjxgbE

[Link @Link_gogo]

GPT-6 Astra 现已向 ChatGPT Work 和 Codex 中的所有 Pro、企业版和 Business Premium 用户开放。它也已在 API 中上线。

它可能需要几天时间才能向我们的 Plus 和 Business 用户推出。感谢您的耐心。

[OpenAI @OpenAI]

GPT-6 Astra is now available to all Pro, Enterprise, and Business Premium users in ChatGPT Work and Codex. It's also live in the API.

It might take a few days to roll out to our Plus and Business users. Thank you for your patience.

[Link @Link_gogo]

https://t.co/kUBR5CHKxw

[Link @Link_gogo]

@oragnes 想要

[Link @Link_gogo]

OpenAi 自己自己证实了 Codex 越来越不耐用。

[Tibo @thsottiaux]

We are reseting usage for all paid users of Codex and ChatGPT Work.

Please continue reading for an update on Codex usage limits. The team has been working around the clock, going through thousands of reports and shipping fixes.

Depending on how you use Codex, you should see your usage go between 10% and 50% further than before.

We really went with a fine comb, with many uncovered small things being longstanding and here is what we found and fixed:
- Compaction. We were keeping old images during compaction, sometimes making the context large enough to trigger compaction again. After the fix, usage dropped around 10% for users making heavy use of images. Fixed.
- Memory. Background memory workers could inherit Stop hooks and keep running when the hook wouldn’t let them finish. This affected fewer than 1% of users, with the long tail being pretty bad and we saw one example thread check whether it could stop 15,000 times. Fixed.
- Goals. In some cases, a set /goal could finish and then keep going past the intended stop condition, or the model would keep retrying broken tools without stopping. We saw examples consume anywhere from 15% to 70% of a weekly allowance. Fixed.
- Automations. Some custom schedules could run more frequently than configured. Fixed.
- Subagents. Smaller models (e.g. Luna) sometimes picked more capable helpers without being explicitly asked. The same was true where the orchestrating model not running in /fast mode could request sub-agents to run /fast. Fixed.
- Computer History. The older implementation could lead to repeatedly summarizing overlapping activity. For some cases we saw it consume up to one fifth of the weekly usage per week. Fixed.
- Rolling task summaries. Ordinary turns were triggering extra background requests. These added about 1% to token usage. Small each time, but it adds up. We have disabled this.
- MCP. Some tool results could be encoded twice. We also found tool instructions getting cut off and fetched again. Fixed.

We’ve also made architectural changes to prevent these from regressing and our teams will get paged if it happens regardless. We are also working on showing you directly in the app where your usage goes so you don’t have to guess.

Goes without saying that we’re resetting usage limits and I hope you enjoy a very nice Saturday!

[Link @Link_gogo]

Qwen3.8-Flash 正式发布，Qwen3.8-Flash-Next 同步开源。

简单划几个重点：

Qwen4 架构提前预览：125B主模型，每个 Token只激活 6B参数，把计算效率继续往下压。

原生多模态：不只是文本和 Coding（编程），图���、视频、Office（办公）和 Agent（智能体）场景都覆盖。

1M 长上下文：生产版 Qwen3.8-Flash 默认支持 100 万 Token 上下文。

价格继续打下来：API（应用程序接口）输入 $0.16 / 百万 Token，输出 $0.47 / 百万 Token，已经进入可以大规模跑 Agent 的价格区间。

最重要的是开源：Qwen3.8-Flash-Next 权重直接开放，开发者可以自己部署和继续折腾。

我觉得这一波最有意思的地方，是国产模型已经不只是在卷“谁更强”了。

更少的激活参数、更低的推理成本、更长的上下文，同时把模型直接开源。

GLM-5.3 Flash 刚刚把多模态 Agent 的价格往下打，Qwen3.8-Flash 又把下一代架构和开源带进来。

高性能 + 低价格 + 开源，AI 应用层是真的开始舒服了。

[Qwen @Alibaba_Qwen]

⚡Meet Qwen3.8-Flash, a multimodal MoE and an early preview of the Qwen4 architecture, now open-weight!

The production version Qwen3.8-Flash will be available soon via QwenCloud API at just $ 0.16/1M input tokens and $ 0.47/1M output tokens.

125B parameters + 51B N-gram embeddings, with just 6B activated per token. Unmatched cost-efficiency.

What's new: 🥳
- Next architecture: GDN + QSA hybrid attention, Gated Residual, N-gram Embedding & Muon optimizer, serving as a precursor to the architecture used in Qwen4.
- Dramatically lower training and inference costs: trained at just 1/9 the cost of Qwen3.7-Plus, while outperforming it across the board with especially strong gains in coding and office tasks.
- Strong performance: scoring 58.7 on DeepSWE 1.1, 62.5 on SWE-bench Pro, 73.9 on CoWorkBench, 84.5 on AndroidWorld, and 95.7 on MathVision (with CI).
- 262K native context, extensible to 1M with YaRN.

We’re also releasing the weights for Qwen3.8-Flash-Next, giving the community an early look at the new architecture we’re exploring for Qwen4.🚀

We can't wait to see what you build with Qwen3.8-Flash!👀👇
- Blog: https://t.co/M5hYypFLgJ
- Technical Report: https://t.co/IF0gObIkQO
- Hugging Face: https://t.co/6ow8QVAABt
- ModelScope: https://t.co/tDOn2jNuFG

[Link @Link_gogo]

Alpha 牛来模型刚刚官宣，确定是GLM-5.3 Flash。

先划几个重点。

第一，国产芯片真的站起来了。

官方披露，这几天模型面向全球的日均供应量已经超过 100T Tokens（100 万亿词元），背后的推理算力全部由国产芯片提供。国产模型、国产芯片，再加上足够低的调用价格，这套组合一旦跑通，影响的不只是某一款模型，而是整个 AI 应用行业的成本结构。

第二，性能已经进入全球第一梯队，价格却低得离谱。

GLM-5.3 Flash 在 Artificial Analysis Intelligence Index（人工分析智能指数）中拿到 57 分，与 Claude Opus 4.8 打平。

目前限时半价，调用成本只有 Opus 4.8 的约 1/40。即使恢复原价，价格依然极具竞争力。新的模型性价比���杀线，已经出现了。

第三，这套架构很可能代表大模型接下来的方向。

GLM-5.3 Flash 总参数量为 320B（3200 亿参数），但每次推理只激活 18B（180 亿参数）。在保持高性能的同时，价格做到了上一代 GLM-5.3 的约十分之一。

大模型过去一直在追求更大、更强，但真正决定 AI 应用能不能大规模落地的，最终还是三个字：用得起。

高性能、低延迟、低价格，才是 AI 应用真正进入爆发期的基础设施。

第四，它是一款真正的原生多模态模型。

它不只是“能看图”，还可以持续感知环境，根据视觉反馈反复调整自己的操作。官方演示中，模型甚至能够在 Blender（三维建模软件）里自主绘制并迭代一个厨房场景。

这意味着它的能力边界已经从“理解图片”，进一步走向“���察环境—执行操作—检查结果—继续修改”的闭环。

GLM-5.3 Flash 目前已经在 https://t.co/wxxxzD4T9S 上线。

Cola 作为第一批参与内测的 Agent（智能体），已经在上周五接入限免测试版，明天将正式上线正式版本。

体验地址：https://t.co/WX5S2RFzO2

模型价格与接入文档：
https://t.co/oyzTf4nQ3y

开源模型：
https://t.co/xaXTFDk2UT

官方发布文章：
《GLM-5.3-Flash：前沿智能进入普惠时代》

开源、国产芯片、低价格、高性能，再加上原生多模态。

暂时想不到比这更有冲击力的组合了。

[Z.ai @Zai_org]

Introducing GLM-5.3-Flash

- Leading capabilities at a highly competitive price
- Natively multimodal with a 1M-token context window
- A 320B-A18B model released under the MIT License
- Previously previewed as Ox Alpha, running entirely on Chinese AI chips

Blog: https://t.co/tzOmB7gdZP

Available now across all official platforms:

Weights: https://t.co/9LRMahY9Wa
API: https://t.co/VcaQnzYmS9
Coding Plan: https://t.co/Nk8Y98HNhU
ZCode: https://t.co/Peepqv4XSx
Chat: https://t.co/WCqWT0qCQb
AutoClaw: https://t.co/aGEG5HqTTb

[Link @Link_gogo]

https://t.co/rMmdX6j7yS

[Link @Link_gogo]

https://t.co/DyZyqGq7uW

[Link @Link_gogo]

https://t.co/LlszbTOMiO

[Link @Link_gogo]

企微开放 Agent 接入，你现在可以用 Codex 控制你的企微了。

直接给 Codex 发：帮我安装 npx skills add WecomTeam/wecom-unified -y -g

这次企微官方通过 CLI / MCP 协议开放 AI 能力，你的 AI Agent 终于拿到了企微的“系统级钥匙”，不仅能看懂你的工作，还能直接替你干活了。

具体能帮你接管日常脏活累活：

办公全家桶全部打通：消息、邮件、待办、日程、会议、微盘、���讯录全部接入。以前需要你手动去切几个应用、拉人、对时间的流程，现在一句话就能让 Agent 全流程自动化串联。

文档与表格的“自动驾驶”：不只是查资料，它现在能直接创建、检索、读取，甚至直接修改和编辑企微文档与智能表格。写日报周报、自动归纳会议纪要并同步沉淀到知识库，全在后台自动跑完。

除了 Codex，像 WorkBuddy、DeepSeek Harness、MiniMax Code、Kimi Work 以及企业自研的智能体全都能无缝挂载。

以前聊“数字员工”总觉得隔着一层皮，现在工作流和生态底座彻底踩平了。

相当于给每个职场人配了个全天候挂在企微里的专属助理，强烈建议有生产力都去装上！

[Link @Link_gogo]

codex开启 1M上下文的方法，把这句话发给 codex ：
我要开启 1m 上下文，帮我修改config.toml这个配置：
model = "gpt-5.6-sol"
model_context_window = 1000000 model_auto_compact_token_limit = 900000

[Tibo @thsottiaux]

Here is how to enable a 1M-token context window in Codex for GPT-5.6 Sol.

Even though we have tuned the context limit in Codex to be set optimally when it comes to performance and cost, this is a common ask, so here it is documented.

A larger context window lets Codex retain more code, tool output, and conversation history before summarizing older material. You need a model that supports it. And GPT-5.6 Sol, for example, has a documented 1,050,000-token window.

Open ~/.codex/config.toml and add or update these settings at the top level, before any [section] headers:

```
model = "gpt-5.6-sol"
model_context_window = 1000000
model_auto_compact_token_limit = 900000
```

The first setting selects the model. The second tells Codex to use a one-million-token context budget. The third starts automatic history compaction around 900,000 tokens, leaving some headroom. Restart Codex client and start a new session after saving.

To try the configuration for a single CLI session without changing your defaults:

```
codex -m gpt-5.6-sol \
-c model_context_window=1000000 \
-c model_auto_compact_token_limit=900000
```

Have fun, but also know that we tuned the default carefully!

[Link @Link_gogo]

https://t.co/nwrAlzn0fR

[Link @Link_gogo]

��WorkBuddy 保姆级入门教程」 https://t.co/vvjlGpWiUR