Guac by the 5 gallon bucket
Margs by the 5 gallon bucket
Best el pastor I ever had
This tailgate they’re putting on at sproul plaza is something special
Piedmont next
@chetaslua Hi everyone! For those asking how Codex interacted with Unity, whether it was 1-shot, etc, I broke down how I built the workflow used to create these assets here: https://t.co/SjiWTJtg1H
A few people have asked how I built the Unity environment in the GPT-6 Astra launch blog. Here’s the workflow:
1. Created a Unity plugin in Codex using Unity’s published skills + the Unity CLI
2. Downloaded city assets from the Unity Asset Store
3. Started a Codex session in the Unity project directory
4. Enabled computer use
5. Gave Codex this prompt:
“Build a new 3D city scene for me. I want several type of buildings (diff heights) and cars there. Also use roads, billboards and other elements that belong to a big city. Everything you need is in this folder Assets/ithappy/Cartoon_City_Free.”
Codex + Astra built the environment using Unity’s skills and CLI, then used computer use to inspect the scene and make adjustments, all in 1 shot!
Happy building!
https://t.co/K3H4jiJp5t
A few people have asked how I built the Unity environment in the GPT-6 Astra launch blog. Here’s the workflow:
1. Created a Unity plugin in Codex using Unity’s published skills + the Unity CLI
2. Downloaded city assets from the Unity Asset Store
3. Started a Codex session in the Unity project directory
4. Enabled computer use
5. Gave Codex this prompt:
“Build a new 3D city scene for me. I want several type of buildings (diff heights) and cars there. Also use roads, billboards and other elements that belong to a big city. Everything you need is in this folder Assets/ithappy/Cartoon_City_Free.”
Codex + Astra built the environment using Unity’s skills and CLI, then used computer use to inspect the scene and make adjustments, all in 1 shot!
Happy building!
https://t.co/K3H4jiJp5t
We did it — the first decentralized training of a 10B model is complete!
Trained across the US, Europe, and Asia 🌐
Post-training with @arcee_ai is underway, and a full open-source release is coming in ~1 week, including: base model, checkpoints, post-trained model and data.