Released: 2026-02-10
Missing Domino:
meta-llama/llama-4-maverick
meta
Reasoning
The excluded domino is number 0 and 0
openai/gpt-5-image-mini
openai
Reasoning
Counting the visible halves for each pip value shows every number 0–6 appears the expected eight times except the 5s, which appear only six times; therefore the missing tile must be the double five (5,5).
x-ai/grok-4-fast
x-ai
Reasoning
By counting the occurrences of each number across all halves of the 27 dominoes, numbers 0,1,3,4,6 appear 8 times each, while 2 and 5 appear 7 times each, indicating the missing domino is [2-5].
google/gemini-2.5-flash-preview-09-2025
Reasoning
A full standard double-six set of dominoes has 28 unique tiles. By identifying the 21 unique tiles visible in the image, we determined that all tiles containing the number ’1’ (i.e., [0, 1], [1, 2], [1, 3], [1, 4], [1, 5], and [1, 6]) are present, with the exception of the double [1, 1]. Although the image contains multiple discrepancies (only 26 tiles are shown, resulting in 7 overall unique tiles missing if no visual identification errors occurred), [1, 1] is the only tile missing that completes the entire sub-set of tiles containing the number 1.
qwen/qwen3-vl-30b-a3b-thinking
qwen
Reasoning
The full set contains 28 dominoes with numbers 0-6. Each number should appear 8 times across all tiles (28 tiles × 2 sides = 56 total dots; 56 ÷ 7 numbers = 8 per number). Counting the occurrences of 6 in the image shows it appears only 6 times (from dominoes (0,6), (1,6), (2,6), (3,6), (4,6), and (5,6)), which is 2 less than the required 8. This discrepancy is explained by the missing double-six domino (6,6), which contributes 2 to the count of 6. All other numbers appear the correct number of times, confirming (6,6) is the missing tile.
anthropic/claude-haiku-4.5
anthropic
Reasoning
By systematically checking all 28 possible dominoes in a standard double-six set, I identified 27 dominoes in the image. The missing domino is [2, 4], which would represent a tile with 2 dots on one end and 4 dots on the other end.
Each model receives the same domino puzzle grid. Models are tasked with identifying the missing domino using only visual clues from the grid. Their answers are evaluated without feedback or retries.