AI vs Puzzles

Back to Home
Puzzle#191

Countdown

Released: 2026-05-06


11
Models Tested
2
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

anthropic/claude-3.7-sonnet

anthropic

682Target
25
2
6
5
9
1

25 * (9 * 2 + 6 + 5 + 1)

= 750

Failed

anthropic/claude-sonnet-4.6

anthropic

682Target
25
2
6
5
9
1

25 * (9 * 6 - 5) + 2 * 1

= 1227

Failed

mistralai/mistral-medium-3

mistral

682Target
25
2
6
5
9
1

((25 * (9 + 5)) * (6 + 1)) + 2

= 2452

Failed

Z.ai

z-ai/glm-5

z-ai

682Target
25
2
6
5
9
1

((25 - 1) * (9 + 6) * 2) + 5

= 725

Failed

google/gemini-2.0-flash-001

google

682Target
25
2
6
5
9
1

((25 * 2) + 6) * (9 + 1)

= 560

Failed

deepseek/deepseek-chat-v3-0324

deepseek

682Target
25
2
6
5
9
1

((25 * (9 + 1)) + 6) * 5

= 1280

Failed

x-ai/grok-3-mini-beta

x-ai

682Target
25
2
6
5
9
1

25 * ((6 * 5) - (2 + 1))

= 675

Failed

google/gemini-3-flash-preview

google

682Target
25
2
6
5
9
1

((25 * 5 + 2) * (6 - 1)) + 9

= 644

Failed

openai/gpt-4o

openai

682Target
25
2
6
5
9
1

((25 * 9) - (6 / 2)) + 5

= 227

Failed

MoonshotAI

moonshotai/kimi-k2.5

moonshot

682Target
25
2
6
5
9
1

(25 + 6) * (9 * 2 + 5 - 1)

= 682

Perfect Solution!

Minimax

minimax/minimax-m2.5

minimax

682Target
25
2
6
5
9
1

(9 * 2 + 5 - 1) * (25 + 6)

= 682

Perfect Solution!

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