AI vs Puzzles

Back to Home
Puzzle#115

Countdown

Released: 2026-02-19


11
Models Tested
5
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

openai/gpt-4o

openai

232Target
25
50
5
10
10
7

(25 * 10) - (50 / 5) + 7

= 247

Failed

google/gemini-2.0-flash-001

google

232Target
25
50
5
10
10
7

(50 * 5) - 10 + 7 - 25

= 222

Failed

Minimax

minimax/minimax-m2.5

minimax

232Target
25
50
5
10
10
7

(25 + 7) * 10

= 320

Failed

deepseek/deepseek-chat-v3-0324

deepseek

232Target
25
50
5
10
10
7

((50 - 10) * 7) + (25 * 5)

= 405

Failed

anthropic/claude-3.7-sonnet

anthropic

232Target
25
50
5
10
10
7

25 * 10 - 50 + 7

= 207

Failed

x-ai/grok-3-mini-beta

x-ai

232Target
25
50
5
10
10
7

((50 * 5) - (25 - 7))

= 232

Perfect Solution!

Z.ai

z-ai/glm-5

z-ai

232Target
25
50
5
10
10
7

50 * 5 - (25 - 7)

= 232

Perfect Solution!

google/gemini-3-flash-preview

google

232Target
25
50
5
10
10
7

(5 * 50) - (25 - 7)

= 232

Perfect Solution!

MoonshotAI

moonshotai/kimi-k2.5

moonshot

232Target
25
50
5
10
10
7

(25 - 7) * 10 + 50 + 10 / 5

= 232

Perfect Solution!

mistralai/mistral-medium-3

mistral

232Target
25
50
5
10
10
7

(25 + 7) * (10 + 10) - 50

= 590

Failed

google/gemini-3.1-pro-preview

google

232Target
25
50
5
10
10
7

(((25 + (10 / 10)) * 7) + 50)

= 232

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