GPT Dots is starting to roll out across the EU
For those unfamiliar, Dots is OpenAI’s proactive assistant, similar to Grok’s bot or a locally running Hermes agent. It can work on its own cloud…
English original
AI Tools & updates. All posts and signals, newest first.
SearchFor those unfamiliar, Dots is OpenAI’s proactive assistant, similar to Grok’s bot or a locally running Hermes agent. It can work on its own cloud…
You can now create a page for your business and get analytics on what customers are searching for in ChatGPT. For now, it's only available for…
Today: Multi-account support in the app Tomorrow: Banked reset Day after tomorrow: Banked reset But we’re still expecting a new model on Day 29 👀
• Redesign the entire UI just by prompting an agent • Let Claude Code and Codex spawn other agents • Get results delivered back automatically •…
Also, we've got another reset ✅
Imagine you have a session with 500K tokens of context. You come back 90 minutes later, your prompt cache has expired, and Claude has to process…
Day 1: WIN Sped up Sol and Astra from ~30 to ~50 tokens per second. Most people use exactly these models, so it's a clear win. Day 2: NOT ACCEPTED…
Claude has prompt caching with a 1-hour TTL, but that’s only for the main agent. Subagents get just 5 minutes by default. If you use long-running…
2.1 is a decent feature. Auto-review for risky actions sounds useful, especially for sub-agents and long-running tasks. But I’m always on bypass…
Going from ~30 TPS to ~50 TPS is already a meaningful improvement and basically puts it at Fable-level speed. That said, I think the bar is higher…
Ripped almost everything out and left only the bare minimum: read, inspect, write. Startup context is now under 2k tokens. I also capped the…
It runs on a heavily modified Pi stack with MiniMax’s own agent runtime and tooling. In FrontierHarness Eval, MiniMax Code hit 76.7% pass rate, 4m…
It works through Claude Code CLI, while most OpenClaw features remain available through MCP: browser, memory, sessions, subagents, cron jobs, etc.…
Before, Remote Control only let you continue a session already running on your computer. Now you can choose your computer and folder, then launch a…
Turn it on and ChatGPT opens straight into Remote, ready to work with Codex on your computer from your phone. How to set it up: ChatGPT → Remote → ⋯…
Origin lets you host repos directly in Cursor, open PRs, browse code, review changes, and run agents inside each repo. You can also connect GitHub…
The project hit 11K stars in just a couple of days. It works in three ways: strips hidden text characters with scripts, disrupts statistical…
It can now work alongside Sol and Terra. Until now, Luna was on the previous subagent system, so users had to rely on workarounds to plug it in.
This applies to long sessions. Previously, opening a huge chat meant Codex loaded almost the entire history at once. Now it will load only the part…
The watermark is created at the model level and becomes part of the text itself. It survives normal copy paste and can still be detected after…
The company updated its safety classifiers and cut false positives by around 85%. Fable 5 will now fall back to Opus 5 far less often for questions…
It now gives tighter, more accurate answers, makes fewer factual mistakes, and feels consistent in both quick and deep thinking modes. Plus and Pro…
Starting next week, it will also come with unlimited text chats and a new Think button for harder questions.
It comes through Open Design, a popular plugin built as an alternative to Claude Design. Your whole team can now work with AI to build and edit…
It runs on the new Muse Spark 1.2. Below is gameplay from a game it built.