AI vs Puzzles

Back to Home
Puzzle#127

Countdown

Released: 2026-03-03


11
Models Tested
5
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

mistralai/mistral-medium-3

mistral

850Target
75
25
50
7
5
6

((75 + 50) * (7 + 5)) + (25 * 6)

= 1650

Failed

google/gemini-2.0-flash-001

google

850Target
75
25
50
7
5
6

(75 + 25) * (7 + 6)

= 1300

Failed

openai/gpt-4o

openai

850Target
75
25
50
7
5
6

(75 + 50 + 25) * 6 + 7 + 5

= 912

Failed

deepseek/deepseek-chat-v3-0324

deepseek

850Target
75
25
50
7
5
6

(75 + 25) * (7 + 6 - 5)

= 800

Failed

anthropic/claude-3.7-sonnet

anthropic

850Target
75
25
50
7
5
6

(75 + 6) * 50 / 5

= 810

Failed

google/gemini-3-flash-preview

google

850Target
75
25
50
7
5
6

(75 * (7 + 5)) - 50

= 850

Perfect Solution!

x-ai/grok-3-mini-beta

x-ai

850Target
75
25
50
7
5
6

75 * (6 + 5) + 25

= 850

Perfect Solution!

Minimax

minimax/minimax-m2.5

minimax

850Target
75
25
50
7
5
6

(75 * 7) + (50 * 6) + 25

= 850

Perfect Solution!

MoonshotAI

moonshotai/kimi-k2.5

moonshot

850Target
75
25
50
7
5
6

(75 + 25) * (7 + 6 - 5) + 50

= 850

Perfect Solution!

Z.ai

z-ai/glm-5

z-ai

850Target
75
25
50
7
5
6

(25 + 5 - 7 - 6) * 50

= 850

Perfect Solution!

anthropic/claude-sonnet-4.6

anthropic

850Target
75
25
50
7
5
6

((75 + 50) * 6) + 25 * (7 - 5)

= 800

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