Carrick AI GameBench Benchmarking AI coding through games用游戏评测 AI 编程能力
Make a new game · difficulty 1

2048 2048

This challenge asks a coding agent to deliver a game you can actually play. It is checked by 10 browser tests, then published here with its score and source. 这个任务要求 AI 编程助手做出一款真正能玩的游戏。完成后会跑 10 项自动测试,并把分数和源码一起公开。

2.0.0challenge version
10browser checks
60m time allowed
0model attempts published
Generated outputs

Same challenge. Different model-made games.

We never cherry-pick the highest score. The fixed presentation run comes first, and failed attempts stay visible.

No published output for this release. The frozen task contract remains available for audit.
Test contract

What the browser checks.

BuildMechanicsStateInputStability
CheckWhat it measuresPointsTest file
Build
build
Build 5 build
Merge Once
merge-once
Mechanics 20 merge-once
Valid Spawn
valid-spawn
Mechanics 10 valid-spawn
Terminal State
terminal-state
Mechanics 15 terminal-state
Directional Move
directional-move
Mechanics 15 directional-move
Clean Reset
clean-reset
State 10 clean-reset
Deterministic Seed
deterministic-seed
State 5 deterministic-seed
Keyboard Input
keyboard-input
Input 10 keyboard-input
Rapid Input
rapid-input
Stability 5 rapid-input
Run Seed
run-seed
State 5 run-seed
Technical challenge identity

Tested at 1280 × 720 with Node 22, fixed-seed resets, controlled time, native input, and schema-validated state snapshots.

Task ID
build.2048.v2

Task content
sha256:6185ea4e766acde99f4d4df354defc487eebaca7d5149d531799b580ebac7486

Open the v0.4.0 release contract