AI vs Puzzles

Back to Home
Puzzle#124

Countdown

Released: 2026-02-28


11
Models Tested
6
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

openai/gpt-4o

openai

979Target
25
100
10
9
4
7

((100 * 9) + 7) + (25 * 4)

= 1007

Failed

google/gemini-2.0-flash-001

google

979Target
25
100
10
9
4
7

(100 * 10) - 25 + 4

= 979

Perfect Solution!

google/gemini-3-flash-preview

google

979Target
25
100
10
9
4
7

10 * (100 - (7 - 4)) + 9

= 979

Perfect Solution!

anthropic/claude-3.7-sonnet

anthropic

979Target
25
100
10
9
4
7

(100 - 4) * 10 - 7

= 953

Failed

deepseek/deepseek-chat-v3-0324

deepseek

979Target
25
100
10
9
4
7

((100 * 9) + 25) + (10 * 4)

= 965

Failed

mistralai/mistral-medium-3

mistral

979Target
25
100
10
9
4
7

((100 + 7) * (4 + 9)) + (25 - 10)

= 1406

Failed

anthropic/claude-sonnet-4.6

anthropic

979Target
25
100
10
9
4
7

((100 - 9) * 10) - 25 + 4

= 889

Failed

x-ai/grok-3-mini-beta

x-ai

979Target
25
100
10
9
4
7

((100 * 10) - (25 - 4))

= 979

Perfect Solution!

Minimax

minimax/minimax-m2.5

minimax

979Target
25
100
10
9
4
7

(100*10)-(25-4)

= 979

Perfect Solution!

Z.ai

z-ai/glm-5

z-ai

979Target
25
100
10
9
4
7

100 * 10 - 25 + 4

= 979

Perfect Solution!

MoonshotAI

moonshotai/kimi-k2.5

moonshot

979Target
25
100
10
9
4
7

100 * 10 - (25 - 4)

= 979

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