AI vs Puzzles

Back to Home
Puzzle#175

Countdown

Released: 2026-04-20


11
Models Tested
5
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

google/gemini-2.0-flash-001

google

242Target
75
2
4
7
10
10

(75 * 4) - (10 + 10) + 2 - 7

= 275

Failed

mistralai/mistral-medium-3

mistral

242Target
75
2
4
7
10
10

(75 * (10 - (7 - (10 / 2)))) + 4

= 604

Failed

openai/gpt-4o

openai

242Target
75
2
4
7
10
10

(75 * 4) - (10 - 2)

= 292

Failed

google/gemini-3-flash-preview

google

242Target
75
2
4
7
10
10

(10 * (75 / (7 - 4))) - 10 + 2

= 242

Perfect Solution!

anthropic/claude-3.7-sonnet

anthropic

242Target
75
2
4
7
10
10

(75 - 2) * 4 - (7 - 10)

= 295

Failed

x-ai/grok-3-mini-beta

x-ai

242Target
75
2
4
7
10
10

((75 * 4) - (10 * 7) + 10 + 2)

= 242

Perfect Solution!

MoonshotAI

moonshotai/kimi-k2.5

moonshot

242Target
75
2
4
7
10
10

(75 - 4) * 2 + 10 * 10

= 242

Perfect Solution!

Minimax

minimax/minimax-m2.5

minimax

242Target
75
2
4
7
10
10

75 * 4 - 7 * 10 + 10 + 2

= 242

Perfect Solution!

deepseek/deepseek-chat-v3-0324

deepseek

242Target
75
2
4
7
10
10

((75 - (10 / (7 + 2))) * 4)

= 295

Failed

Z.ai

z-ai/glm-5

z-ai

242Target
75
2
4
7
10
10

(10 + 10 + 2) * (7 + 4)

= 242

Perfect Solution!

anthropic/claude-sonnet-4.6

anthropic

242Target
75
2
4
7
10
10

75 * (2 + 10/10) + 4 * 7

= 253

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