AI vs Puzzles

Back to Home
Puzzle#183

Countdown

Released: 2026-04-28


10
Models Tested
0
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

mistralai/mistral-medium-3

mistral

586Target
25
100
75
3
2
1

((100 + 25) * (75 / 3)) + 1

= 3126

Failed

openai/gpt-4o

openai

586Target
25
100
75
3
2
1

((75 + 25) * 3) + (100 + 1)

= 401

Failed

google/gemini-3-flash-preview

google

586Target
25
100
75
3
2
1

((100 + 75 + 1) * 3) + 2 * 25

= 578

Failed

deepseek/deepseek-chat-v3-0324

deepseek

586Target
25
100
75
3
2
1

(100 * (75 - (25 - 3))) - (2 * 1)

= 5298

Failed

x-ai/grok-3-mini-beta

x-ai

586Target
25
100
75
3
2
1

(75 * (3 * 2 + 1)) + (100 - 25)

= 600

Failed

Minimax

minimax/minimax-m2.5

minimax

586Target
25
100
75
3
2
1

(100+75)*3+25*2+1

= 576

Failed

anthropic/claude-3.7-sonnet

anthropic

586Target
25
100
75
3
2
1

(3 * 75) + (100 * 2) + 25 + 1

= 451

Failed

MoonshotAI

moonshotai/kimi-k2.5

moonshot

586Target
25
100
75
3
2
1

(100 * 2 * 3 - 75 / 25 - 1)

= 596

Failed

Z.ai

z-ai/glm-5

z-ai

586Target
25
100
75
3
2
1

(100 - 2) * (75 / 25 + 3) - 1

= 587

Close; 1 away

anthropic/claude-sonnet-4.6

anthropic

586Target
25
100
75
3
2
1

(100 + 75 + 3) * 2 * (25 - 1)

= 8544

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