Selva

Data Platform Services Engineer

[Thariq @trq212]

We're adding support for AGENTS.md to Claude Code.

Starting today in version 2.1.277, if there is no CLAUDE.md in a folder, Claude will check for and use AGENTS.md.

You can toggle this behavior in /config.

[Selva @Learndecraft]

@Taniyatweets_ It’s the sweetest spot for me after seeing all the demand and its impact on pricing 😂

[Selva @Learndecraft]

@HermesWatcher Telegram easy to configure and work while traveling

[Institute of Foundation Models @IFM_AI]

Introducing K2 Horizon: a connected fleet of six foundation models ranging from 0.9 billion to 375 billion parameters.

- Frontier performance: Across coding and agentic tasks, K2 Horizon delivers top-tier performance in every size class—with the 0.9B, 3.7B and 7B models setting new state of the art at their respective scales.
- Radical openness: K2 Horizon represents the largest fully open-source model launch in AI history. The fully open code, training data and recipes are a significant step forward in transparency.

Launch page: https://t.co/gg0k803SbL
Tech blog: https://t.co/g35L5xMGdS
Hugging Face: https://t.co/3Lb28JhyG9

[Selva @Learndecraft]

@HermesWatcher A personal coach, a personal assistant . A work buddy / employee to chat with and build solutions for problems

[Selva @Learndecraft]

One-founder AI startup org: 8 bots (CEO→CFO→CTO→Analyst→Critic→Dev→QA→Deploy) as Hermes profiles. A red-team bot attacks every report before you see it. Nothing ships without your https://t.co/4gG4JfaRkI: SOULs + doctrine + installer →
https://t.co/LZY8gEDZ2T
#hermes #ai

[Selva @Learndecraft]

@Teknium Yessss

[Selva @Learndecraft]

@HermesWatcher I setup an organization structure with bots like CEO, CFO , Dev, QA etc. it’s fun and fullfiling to see the ideas being worked across the org on Kanban. But Bots are duplicates of my profiles .

[Google Research @GoogleResearch]

Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: https://t.co/CDSQ8HpZoc

[hoeem @hooeem]

https://t.co/4tchgZa4oC

[Yun-Ta Tsai @yunta_tsai]

Technical presentation from Ashok 👇

[Andrej Karpathy @karpathy]

Excited to release new repo: nanochat!
(it's among the most unhinged I've written).

Unlike my earlier similar repo nanoGPT which only covered pretraining, nanochat is a minimal, from scratch, full-stack training/inference pipeline of a simple ChatGPT clone in a single, dependency-minimal codebase. You boot up a cloud GPU box, run a single script and in as little as 4 hours later you can talk to your own LLM in a ChatGPT-like web UI.

It weighs ~8,000 lines of imo quite clean code to:

- Train the tokenizer using a new Rust implementation
- Pretrain a Transformer LLM on FineWeb, evaluate CORE score across a number of metrics
- Midtrain on user-assistant conversations from SmolTalk, multiple choice questions, tool use.
- SFT, evaluate the chat model on world knowledge multiple choice (ARC-E/C, MMLU), math (GSM8K), code (HumanEval)
- RL the model optionally on GSM8K with "GRPO"
- Efficient inference the model in an Engine with KV cache, simple prefill/decode, tool use (Python interpreter in a lightweight sandbox), talk to it over CLI or ChatGPT-like WebUI.
- Write a single markdown report card, summarizing and gamifying the whole thing.

Even for as low as ~$100 in cost (~4 hours on an 8XH100 node), you can train a little ChatGPT clone that you can kind of talk to, and which can write stories/poems, answer simple questions. About ~12 hours surpasses GPT-2 CORE metric. As you further scale up towards ~$1000 (~41.6 hours of training), it quickly becomes a lot more coherent and can solve simple math/code problems and take multiple choice tests. E.g. a depth 30 model trained for 24 hours (this is about equal to FLOPs of GPT-3 Small 125M and 1/1000th of GPT-3) gets into 40s on MMLU and 70s on ARC-Easy, 20s on GSM8K, etc.

My goal is to get the full "strong baseline" stack into one cohesive, minimal, readable, hackable, maximally forkable repo. nanochat will be the capstone project of LLM101n (which is still being developed). I think it also has potential to grow into a research harness, or a benchmark, similar to nanoGPT before it. It is by no means finished, tuned or optimized (actually I think there's likely quite a bit of low-hanging fruit), but I think it's at a place where the overall skeleton is ok enough that it can go up on GitHub where all the parts of it can be improved.

Link to repo and a detailed walkthrough of the nanochat speedrun is in the reply.

[Selva @Learndecraft]

Much needed !

[Noah Cat @Cartidise]

whoever at Google invented this feature, i love you

[Morgan @morganlinton]

After using @diabrowser for a couple of weeks I can confirm, this isn't just another browser, it's a paradigm shift in how we use the Internet.

I don't think I can go back to using a traditional browser ever again.

Here's a good example of how I use Dia, daily. A 🧵

[Ben Lang @benln]

Sam Altman's guide on how to be productive:

[Aadit Sheth @aaditsh]

This guy literally made Cursor 10x more useful with this one system

[Elon Musk @elonmusk]

How can this be real?

[Raspberry Pi @Raspberry_Pi]

��� New product 🚨 Raspberry Pi 400 is your complete personal computer, built into a compact keyboard. Available today from $70. https://t.co/YeUkamOHYc

[Selva @Learndecraft]

Fall Foliage Map 2020 & Nationwide Peak Leaf Forecast: https://t.co/jYb5IB9Gxl via @vacationhere

[Selva @Learndecraft]

@QuinnyPig Tabs! four spaces !