
Jev vs Laya: the same model deciding, at ten times the speed
Jev vs Laya. Same board, same food, same decisions. One ate 47. The other ate 4.
OpenOne brief to two AI agents: build a 3D Rick Sanchez out of nothing but code.
One brief to two AI agents: build a 3D Rick Sanchez out of nothing but code. No model files, no images, no textures. The spiky blue-grey hair, the unibrow, the lab coat over the blue shirt, the drool and the flask, all from shapes each model wrote itself, turning on a turntable in one HTML file.
Claude Fable 5.1 ran in Claude Code, GPT-6 Astra in Codex CLI. One attempt each, no retries. In the reel both figures turn exactly once, so you are comparing the models and not the turntable speeds.
Fable lit a single figure on a dark background, with the flask, the drool and the blue shirt, and let the model carry the frame. Astra put its figure on a turntable inside an 'interdimensional archive' card, C-137 in the corner. Both are recognisable, and neither loaded an asset.
Fable billed $2.99 over about 7 minutes. Astra billed $0.62 in under 3. They ran in different tools, so part of that gap is the tool and not the model.
| Entrant | Tool | Time | Cost | Verdict |
|---|---|---|---|---|
| Claude Fable 5.1 | Claude Code | 7m09 | $2.99 | Single lit figure on a dark stage |
| GPT-6 Astra | Codex CLI | 2m54 | $0.62 | Figure in an archive card |
The reels are the test bench. The real work is picking the one job in your business worth handing to AI, building it properly, and keeping it running. That’s what I do as your AI partner.

Jev vs Laya. Same board, same food, same decisions. One ate 47. The other ate 4.
Open
10,000 Skittles, sorted by colour one at a time.
Open
Four Claudes. One airport.
Open