Ray

Building my startup.

[Ray @0smboy]

受karpathy启发利用notebookLM+ideogram在Spotify for Podcasters搭建了一套个人MBA播客课程👇
#podcast #Spotify #notebookLM #播客 #MBA

[Ray @0smboy]

good work

[Migel Tissera @migtissera]

I'm releasing Tess-4-9B.

Tess-4-9B is the compact sibling of Tess-4-27B, built on the Qwen/Qwen3.5-9B-Base model.

It was post-trained on the same deliberate blend as the 27B release: 64K-token long-context agentic traces — real engineering work done with Fable-5, not synthetic generations — with a reasoning style approximated from Fable-5 by a three-model teacher ensemble (Opus-4.8, GPT-5.5, and GLM-5.2) fused into one coherent voice.

Significant improvement on all measured axes against Qwythos-9B.

[PrismML @PrismML]

Today, we’re announcing Bonsai 27B: the first 27B-class model to run on a phone.

Bonsai 27B is the new multimodal flagship of the Bonsai family. Based on Qwen3.6 27B, it brings a new capability tier to local AI: multi-step reasoning, structured tool use, long-context workflows, and coherent agentic loops.

Until now, models in this class have been impractical to deploy locally. A 27B model occupies roughly 54 GB in 16-bit precision, and even a strong 4-bit build is around 18GB - too large for a phone and for most laptops.
Bonsai 27B changes that.

It comes in two variants:
• Ternary Bonsai 27B: 5.9 GB, 1.71 effective bits per weight, optimized for laptop-class quality.
• 1-bit Bonsai 27B: 3.9 GB, 1.125 effective bits per weight, optimized for phone-class footprint.

Everything is open-sourced today under the Apache 2.0 license.

[Palantir @PalantirTech]

Anyone can build an agent. But to build a trustworthy agent at enterprise scale that is durable, long-running, optimized, contextually aware, and autonomous, the right infrastructure is required.

At DevCon 6 we introduced the Agent Stack, the culmination of learnings gathered over years of agentic implementations. Orchestrator, Agent Engine, Agent SDK, Agent Builder, Agent Manager, AIP Evolve, SuperRepo, and so much more.

All built on the Ontology, to power agents that actually work in production.

[Xenova @xenovacom]

Bonsai 27B just changed the local LLM game forever.

1-bit quantization shrinks it from 54GB to just 3.8GB (-93%), while retaining 90% of its intelligence. That's insane.

With custom WebGPU kernels written by Fable 5 and GPT 5.6 Sol, the model now runs locally in your browser!

[Richard Sutton @RichardSSutton]

I can’t say enough good things about John Carmack @ID_AA_Carmack and his Keen Technologies. But now Khurram Javed @kjaved_ and I have broken away to start our own startup and pursue a slightly different path toward understanding intelligence. Like Keen (and like Ineffable) we at Oak Lab @oaklab_ai believe in reinforcement learning and that intelligence is created and maintained from run-time experience. But we think current deep learning methods are weak and inefficient, and need not more tweaks, but fundamentally new ideas and a thorough reworking before they can provide a solid foundation for achieving the more ambitious goals of AI.

[Ray @0smboy]

@skaas777 还有GPT live

[Ray @0smboy]

我的飞轮系统，哈哈😆

[Ray @0smboy]

我构建了一个codex模型自动选择器，它会根据你的问题自动选择模型来回答

[Ray @0smboy]

@mattshumer_ 我之前自己不小心手动干过同样的事，后来Claude code帮我全恢复回来了，恢复的过程我记录了这个skill：https://t.co/VJn29GqClM 可能会帮到你，减少一些损失

[Ray @0smboy]

free url deploy

[Logan Kilpatrick @OfficialLoganK]

Today we are rolling out pretty URL’s for deployed apps in @GoogleAIStudio, each app can get a “https://t.co/FvJdcNxK0w”, for free!

Free apps, free deploys, and now free pretty URLs!

[Ray @0smboy]

@MetaforDevs region is your best result

[Ray @0smboy]

@Huanusa https://t.co/HMU849LKOc，有对应的单词游戏网站，大家可以自由学习

[Ray @0smboy]

train a model from one prompt using GPT5.6

[Pietro Schirano @skirano]

You can now vibe code a language model.

From a single prompt, GPT‑5.6 built the entire training pipeline and trained a model from scratch on my iMessage history. Locally on my Mac.

It now generates replies in my writing style.

[Ray @0smboy]

what？？？

[Mark Zuckerberg @finkd]

(1) Today we're releasing Muse Spark 1.1 -- a strong agentic and coding model at a very low price. It's available through our new Meta Model API and in Meta AI.

[Ray @0smboy]

我也看好Google Gemini

[Ryan Lopopolo @_lopopolo]

Some personal news: I’m joining @googlecloud as Principal Engineer, Agentic Google Cloud Platform.

I am excited to chase after building the future of work as part of a cloud that deeply adapts to customer knowledge and workflows and environments.

@OpenAI was the first place in my career where “you can just do things” was the real and honest truth. I am very grateful for the run, the people, and the weird magical stretch of getting to help make agents real.

Let’s go fam, plenty of AGI out there to build for.

[Ray @0smboy]

amazing

[Elon Musk @elonmusk]

Cool

[Ray @0smboy]

@rien7z 会给你保留，但是会降智，所以我立马续费了

[Ray @0smboy]

someone get it

[Daichi Konno / 紺野 大地 @_daichikonno]

GPT-5.6-Solキターーーー！！！🎉

みなさんの所にも、もう来ましたか？それともちょっとだけ先行公開？😳

色々と試していきます！！🔥

[Ray @0smboy]

我的底气来源：
1，使用最好的模型和 agent，信任他们；
2，使用我的个人偏好模型（已训练好并持续迭代中），让它生成我喜欢的回答模式；
3，使用我的个人决策模型（已训练好并持续迭代中），在模型和 agent 之间替我做一些日常简单的决策，重要决策由我本人来定夺，候选来自top 模型和个人决策模型推荐；
4，我的决策和回复会生成偏好语料和决策语料反过来喂养迭代偏好模型和决策模型；
5，我主要的任务就是每天冒泡新的想法，派发任务，查看任务进度和质量，其他的都交给我的主理人系统来完成，主理人系统包括多 agent 子群和个人偏好和决策模型模型组成

[AlexZ 🦀 @blackanger]

我的底气来源：

1. 使用最好的 Coding 模型 和 agent，然后信任它
2. 使用 Rust 语言，因为 Rust 编译器可以为 AI 生成的代码兜底。这个兜底的意思是，它能保证代码的正确性。
3. 使用 agent-spec （我自己实现的 sdd 工具），使用 Spec 驱动开发方式，把我的意图“编译” 为 Spec 文件（包含了意图/约束/验证方法）
4. 人工测试，是的，我从一个正经开发沦为了测试人员
5. 我近 20 年的开发经验，如果 ai 定位不到问题，我会帮助它 debug

[Ray @0smboy]

big step

[Elon Musk @elonmusk]

Artificial Analysis assessment