In honor of GPT-6 community night, here's some of the coolest projects built by the community Shoutout to everyone who's been sharing what they've built! (edited by Astra, obv)

In case you missed it, @pvncher wrote a blog on cleaning up your skills and agents.md to better fit Astra my advice? let Codex audit and update them for you

Astra is very good at 3D modeling, and I can't wait for all of you to experience it, for now here is a little walkthrough on how I built the demo house for our launch blog post. From a Blender scene to a Unreal Engine 5 walkable experience. 🧵

after a year as an independent creator, i'm joining @openai imagining the future is how we create it we need to expand who gets to imagine and what becomes possible. those are the stories i’m here to tell if you're an artist, storyteller or maker, i'd love to hear from you 🫶🏼

I switched my Hermes agent to 100% free local models with Magnitude It’s running Qwen 3.6 35B-A3B at ~60 tok/s on my DGX Spark. It’s free, private, and always on running background tasks Hermes set Magnitude up itself. With the CLI it: - Profiled my hardware and found the best models for it - Walked me through the options and let me decide - Switched itself over From there, models load just in time as the agent works and unload when idle Works with Pi, OpenCode, Hermes, OpenClaw, Codex, Claude Code, Oh My Pi, and Cline Copy this prompt and send it to your agent: “Set up local models for me with the Magnitude CLI. Install it with `npm i -g @magnitudedev/cli` (or my package manager), then run `magnitude docs onboarding` and follow the instructions” GitHub:

Cost per task & time per task really need to be the measurements going forward I’m convinced any model can produce brilliant outputs given enough test-time compute Give me unlimited time and I might have a sliver of a chance at world-championship chess, but that’s not the setup, for a reason

Astra is significantly more token efficient and more capable at finding vulnerabilities and developing exploits. We’re planning to make it available soon, with access to its most advanced cyber capabilities limited at first, then expanding over time.

Something I’ve been thinking about: We need to consider how we represent model advancements via demos in terms of economic productivity, not just flashy visuals/games I’m absolutely guilty of the former; games can be eye popping and quickly demonstrate abilities in terms that many people understand. But given the scrutiny our industry is under, does this best represent what is possible?

Worried that AI is going to ruin the sanctity of Fantasy Leagues In college, we almost banned one of our friends because he had access to ESPN+ There's no way to beat the guy who has Codex drafting for him and managing his team all season long