A new, more powerful version of Anthropic's Mythos has come out of training. In itself, this is nothing extraordinary. What else could one expect? That Mythos is already the end? Of course not. This is only the beginning. What is exciting here is the
SOFTWARE
-
Realtime voice with pets heavy on UI thread, CDP DOM manipulation issues
By
–
Realtime voice with pets is a little bit too heavy on the UI thread for me. I have to force quit quite often. Also DOM manipulation with CDP on browser use makes codex stuck if I ask any follow-up questions to operate it further.
-
Comparing models is not limited to pure performance
By
–
Comparing models is not just a matter of pure performance.
-
Issues with Codex and Code: division, lack of exploration and testing
By
–
This is aside from the other key "software brain" problems of Codex and Code: dividing all work into front-end and back-end design, solving for the general case in a repeatable way, not testing or exploring idea spaces, testing for technical correctness but not other aspects…
-
Software-brained approach limits code tools for knowledge work
By
–
A fundamental problem with extending Codex/Cowork/Code to all knowledge work is that they remain very "software-brained" where the end result (the software) is what is important & that code serves as a source of truth. For a lot of other knowledge work, the process is at least
-
New Product Raven: Self-Evolving Agent OS
By
–
我又来剧透了
— 艾略特 (@elliotchen100) 21 juin 2026
从龙虾到爱玛氏，从爱玛氏到什么？
没错，我们马上要发布一个全新的产品：Raven。
它不是又一个「会调用很多工具」的 Agent。
它是一个会自我进化的 Agent OS。
大多数 Agent 的学习，停在技能层：多学一个工具，多记一条流程，多写一段 prompt。
Raven 不一样。… pic.twitter.com/KoO8KGMrPCHere I go again with a spoiler. From Lobster to Aimashi, from Aimashi to what? That's right, we are about to release a brand new product: Raven. It's not just another Agent that 'can call many tools'. It's a self-evolving Agent OS. Most Agents' learning stops at the skill level: learn one more tool, memorize one more process, write one more prompt. Raven is different.
-
Turn any paper into running code with autoarxiv
By
–
Turn any paper into running code.
— Akshay 🚀 (@akshay_pachaar) 21 juin 2026
Just swap arxiv → autoarxiv in the paper url.
That hands the paper to an AI agent from alphaXiv. It reads the abstract, the claims, and the linked GitHub repo, then clones the codebase and works through the usual setup pain like dependencies,… pic.twitter.com/UOPJnWdfLJTurn any paper into running code. Just swap arxiv → autoarxiv in the paper url. That hands the paper to an AI agent from alphaXiv. It reads the abstract, the claims, and the linked GitHub repo, then clones the codebase and works through the usual setup pain like dependencies,
-
Reference to GPT-5.6’s superiority in front-end
By
–
This is a reference to the fact that GPT-5.6 is significantly better in front-end, isn't it?
-
Meta AI unveils Artifacts tab to store presentations and documents
By
–
Meta AI gets a new Artifacts tab on the web. All presentations, documents, web pages and other creations would be stored there. Bridging the gap.
-
Vercel CEO impressed by GLM-5.2, open source and open weights
By
–
Even the CEO of Vercel is impressed/shocked by the exceptional performance of GLM-5.2 in coding. open source, open weights.