AI vs Puzzles

Back to Home
Puzzle#145

Countdown

Released: 2026-03-21


11
Models Tested
3
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

google/gemini-2.0-flash-001

google

660Target
100
75
50
5
10
7

(100 - 10) * 7 + 50 - 5

= 675

Failed

deepseek/deepseek-chat-v3-0324

deepseek

660Target
100
75
50
5
10
7

(75 - 5) * (10 + 7)

= 1190

Failed

google/gemini-3-flash-preview

google

660Target
100
75
50
5
10
7

(100 + 10) * (7 - (50 / 75))

= 696

Failed

Minimax

minimax/minimax-m2.5

minimax

660Target
100
75
50
5
10
7

(100 * 7) + (75 - 10) + 5

= 770

Failed

x-ai/grok-3-mini-beta

x-ai

660Target
100
75
50
5
10
7

(100 * 7) - (50 - 10)

= 660

Perfect Solution!

Z.ai

z-ai/glm-5

z-ai

660Target
100
75
50
5
10
7

(100 * 7) - 50 + 10

= 660

Perfect Solution!

MoonshotAI

moonshotai/kimi-k2.5

moonshot

660Target
100
75
50
5
10
7

7 * 100 - 50 + 10

= 660

Perfect Solution!

openai/gpt-4o

openai

660Target
100
75
50
5
10
7

100 * 5 + 75 + (50 / 10)

= 580

Failed

anthropic/claude-3.7-sonnet

anthropic

660Target
100
75
50
5
10
7

(100 - 10) * 75 / 5

= 1350

Failed

anthropic/claude-sonnet-4.6

anthropic

660Target
100
75
50
5
10
7

75 * (10 - 5) + 100 + 50 + 7

= 532

Failed

mistralai/mistral-medium-3

mistral

660Target
100
75
50
5
10
7

(100 + 75 - 10) * (7 + 5)

= 1980

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