AI vs Puzzles

Back to Home
Puzzle#151

Countdown

Released: 2026-03-27


11
Models Tested
4
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

mistralai/mistral-medium-3

mistral

350Target
100
75
5
10
8
6

(100 + 75) * (10 - (8 - 6))

= 1400

Failed

google/gemini-2.0-flash-001

google

350Target
100
75
5
10
8
6

(100 * 5) - 75 + 10 - 6 - 8

= 421

Failed

anthropic/claude-3.7-sonnet

anthropic

350Target
100
75
5
10
8
6

(100 * 5) - 75 - (10 - 8 - 6)

= 429

Failed

anthropic/claude-sonnet-4.6

anthropic

350Target
100
75
5
10
8
6

75 * 5 - 100 * (10 - 8 - 6)

= 775

Failed

deepseek/deepseek-chat-v3-0324

deepseek

350Target
100
75
5
10
8
6

((100 + 6) * (10 + 5)) - (75 + 8)

= 1507

Failed

Z.ai

z-ai/glm-5

z-ai

350Target
100
75
5
10
8
6

(75 - 5 * 8) * 10

= 350

Perfect Solution!

Minimax

minimax/minimax-m2.5

minimax

350Target
100
75
5
10
8
6

(100 - 75 + 6) * 10

= 310

Failed

MoonshotAI

moonshotai/kimi-k2.5

moonshot

350Target
100
75
5
10
8
6

(100 + 75) * (8 - 6)

= 350

Perfect Solution!

openai/gpt-4o

openai

350Target
100
75
5
10
8
6

(75 * 5) - ((10 - 6) * 8)

= 343

Failed

google/gemini-3-flash-preview

google

350Target
100
75
5
10
8
6

(100 + 75) * (10 / 5)

= 350

Perfect Solution!

x-ai/grok-3-mini-beta

x-ai

350Target
100
75
5
10
8
6

((100 + 75) * (10 / 5))

= 350

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