AI vs Puzzles

Back to Home
Puzzle#162

Countdown

Released: 2026-04-07


11
Models Tested
3
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

anthropic/claude-3.7-sonnet

anthropic

789Target
100
25
3
7
3
9

100 * 7 + 25 * 3 + 9 + 3

= 787

Close; 2 away

openai/gpt-4o

openai

789Target
100
25
3
7
3
9

(100 * 7) + (3 * 25) + (3 * 9)

= 802

Failed

google/gemini-2.0-flash-001

google

789Target
100
25
3
7
3
9

(9 * (100 - 3)) - (25 + 7 + 3)

= 838

Failed

google/gemini-3-flash-preview

google

789Target
100
25
3
7
3
9

(3 * ((3 * 100) - (25 + 9 - 7)))

= 819

Failed

deepseek/deepseek-chat-v3-0324

deepseek

789Target
100
25
3
7
3
9

(100 - (25 - (9 - 3))) * (7 + 3)

= 810

Failed

anthropic/claude-sonnet-4.6

anthropic

789Target
100
25
3
7
3
9

( 100 * 7 ) + ( 25 * 3 ) + ( 9 - 3 )

= 781

Failed

x-ai/grok-3-mini-beta

x-ai

789Target
100
25
3
7
3
9

(100 * 7) + (25 * 3) + 9 + 3

= 787

Close; 2 away

MoonshotAI

moonshotai/kimi-k2.5

moonshot

789Target
100
25
3
7
3
9

3 * (7 * 25 + 100 - 9 - 3)

= 789

Perfect Solution!

Minimax

minimax/minimax-m2.5

minimax

789Target
100
25
3
7
3
9

(100 + 9) * 7 + 25 + 3/3

= 789

Perfect Solution!

Z.ai

z-ai/glm-5

z-ai

789Target
100
25
3
7
3
9

(100 + 9) * 7 + 25 + 3 / 3

= 789

Perfect Solution!

mistralai/mistral-medium-3

mistral

789Target
100
25
3
7
3
9

(100 + 25) * (9 + 3) + (7 * 3)

= 1521

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