AI vs Puzzles

Back to Home
Puzzle#165

Countdown

Released: 2026-04-10


11
Models Tested
4
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

anthropic/claude-3.7-sonnet

anthropic

475Target
100
25
1
7
2
8

100 * (2 + 8) - 25

= 975

Failed

mistralai/mistral-medium-3

mistral

475Target
100
25
1
7
2
8

((100 + 25) * (8 - (7 - (2 / 1))))

= 375

Failed

deepseek/deepseek-chat-v3-0324

deepseek

475Target
100
25
1
7
2
8

(100 - (25 + 1)) * 7

= 518

Failed

Minimax

minimax/minimax-m2.5

minimax

475Target
100
25
1
7
2
8

100 * (7 - 2) - 25

= 475

Perfect Solution!

x-ai/grok-3-mini-beta

x-ai

475Target
100
25
1
7
2
8

(100 * (7 - 2)) - 25

= 475

Perfect Solution!

MoonshotAI

moonshotai/kimi-k2.5

moonshot

475Target
100
25
1
7
2
8

100 * (7 - 2) - 25

= 475

Perfect Solution!

google/gemini-2.0-flash-001

google

475Target
100
25
1
7
2
8

(100 * 2) + 25 + 7 + 8 + 1

= 241

Failed

openai/gpt-4o

openai

475Target
100
25
1
7
2
8

(8 - 1) * (100 - 25) + 7

= 532

Failed

google/gemini-3-flash-preview

google

475Target
100
25
1
7
2
8

(100 - 2 - 25) * 7 + 1 - 8

= 504

Failed

Z.ai

z-ai/glm-5

z-ai

475Target
100
25
1
7
2
8

100 * (7 - 2) - 25

= 475

Perfect Solution!

anthropic/claude-sonnet-4.6

anthropic

475Target
100
25
1
7
2
8

100 * (2 + 7) / (1 + 8 - 25)

= -56

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