Benchmark summary
The model builds 2048, the sliding tile puzzle in SC# as an agent: it writes the app, compiles it, runs its own tests and fixes the code. Hidden tests check the rules from the task; each hidden test is its own result.
Tests with multiple category tags are counted in each matching category, so category totals may exceed the report total.
| Benchmark items | |||
|---|---|---|---|
| Script / test | Status | Tokens | |
Building Games in SC#: 2048
11 passed |
100.0% |
IN1 504 642 OUT14 020 | |
| ✓ | Game2048The model builds the game 2048 from the task as an agent; this item passes when the app compiles and can be exported, and each hidden test is its own item.1 AI query
36 tool calls |
Passed | IN1 504 642 OUT14 020 |
| ✓ | Hidden_BestScoreThe best score follows the score and stays for the next game.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_GameOverWhen no move is left after a move, the game is over; arrows do nothing then, and Enter starts a new game and does nothing else in that frame.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_InitialStateAfter AppInit the grid has exactly two tiles (2 or 4) and the counters have their starting values.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_KeyPressesHolding an arrow makes one move; only a new press makes the next one. With two arrows at once, Left, Right, Up, Down is the order.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_ManyMovesEvery move that changes the grid gives the slide result plus exactly one new tile on a cell that was left empty.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_MoveLeftLeft slides and merges every row from the left, each tile merging at most once, and adds the merged values to the score.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_MoveRightUpDownRight, Up and Down slide and merge from their own side.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_NoChangeNoTileA move that changes nothing adds no tile, no move and no score.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_ScreenThe screen shows the board, the cells in the colors of their values, and the texts.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |
| ✓ | Hidden_WinningMaking a 2048 tile wins: the win screen ignores arrows until Enter, then the game goes on and does not show the win screen again.0 AI queries
0 tool callsHiddenTest |
Passed | IN0 OUT0 |