AI vs Puzzles

Back to Home
Puzzle#248

Countdown

Released: 2026-07-02


8
Models Tested
0
Solved
8
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

openai/gpt-4o

openai

599Target
100
50
75
9
4
7

((100 + 9) * 7) - (75 + 4)

= 684

Failed

mistralai/mistral-medium-3

mistral

599Target
100
50
75
9
4
7

((100 + 50) * (9 - 4)) + (75 / 7)

= 760

Failed

deepseek/deepseek-chat-v3-0324

deepseek

599Target
100
50
75
9
4
7

(100 * (7 - (50 / (75 - 9)))) + 4

= 628

Failed

Minimax

minimax/minimax-m2.5

minimax

599Target
100
50
75
9
4
7

(100+50)*4 - 9 + 7

= 598

Close; 1 away

Z.ai

z-ai/glm-5

z-ai

599Target
100
50
75
9
4
7

(75 + 50) * 4 + 100 + 7 - 9

= 598

Close; 1 away

anthropic/claude-sonnet-4.6

anthropic

599Target
100
50
75
9
4
7

75 * (9 - 4 + 7) - 50 - 100

= 750

Failed

MoonshotAI

moonshotai/kimi-k2.5

moonshot

599Target
100
50
75
9
4
7

4 * (100 + 50) - 9 + 7

= 598

Close; 1 away

google/gemini-3-flash-preview

google

599Target
100
50
75
9
4
7

((75 + 9) * 7) + 50 - 4

= 634

Failed

Methodology Note

Each model receives the same prompt with the numbers to use. Models are tasked with creating an expression using only arithmetic operations to reach the target number. Each number can only be used once and you do not have to use all the numbers. Their answers are evaluated without feedback or retries.

Leaderboard