Call of Duty in the browser — Claude Code vs Codex CLI on one prompt
mshumer's three-paragraph prompt asks for a AAA first-person shooter in Three.js and never says when to stop: fan out sub-agents, loop until a harsh critic is wowed. Kimi K3 under Codex CLI finished in 74 minutes for $3.68. Opus 5 under Claude Code was interrupted at the 90-minute mark having spent $18.57 — its build still boots, but takes 62 seconds to load its world before the menu appears.
"Call of Duty in the browser — Claude Code vs Codex CLI on one prompt" is one prompt run by 2 models: Opus 5 and Kimi K3.1 of 2 finished; 2 measured locally.
Models
2
Completed
1 of 2
Cheapest completed run
Kimi K3 — $3.68
Measured locally in Bench Arena: tokens reconciled on the harness log, cost recomputed from them.
Output
Opus 5
Kimi K3
Prompt6 lines · EN
I want you to build a first-person shooter at the level of the most recent Call of Duty games. It should be utterly perfect, visually beautiful, with every single thing done at AAA quality—from textures to physics to anything you could think of.
Fan out sub-agents and have sub-agents tackle each one individually so that the game is utterly perfect. You should /loop on each item and have a separate sub-agent check it visually to ensure it looks triple A. That separate sub-agent should be a really harsh critic, and if it doesn't look triple A, it should keep going.
Don't stop until each sub-agent is utterly wowed with the quality when compared with the actual Call of Duty game. It should literally compare them side by side blind and say which one looks better. Do this in ThreeJS. /loop until it's utterly perfect. Fan out sub-agents and ultracode.
