AI vs Puzzles

Back to Home
Puzzle#143

Countdown

Released: 2026-03-19


10
Models Tested
0
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

mistralai/mistral-medium-3

mistral

414Target
25
5
5
10
10
5

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

= 400

Failed

anthropic/claude-3.7-sonnet

anthropic

414Target
25
5
5
10
10
5

(10 * 5 * 5 + 25) * 10 / 5

= 550

Failed

deepseek/deepseek-chat-v3-0324

deepseek

414Target
25
5
5
10
10
5

(25 * (10 + 5)) + ((10 + 5) * 5)

= 450

Failed

google/gemini-2.0-flash-001

google

414Target
25
5
5
10
10
5

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

= 390

Failed

x-ai/grok-3-mini-beta

x-ai

414Target
25
5
5
10
10
5

(25 * (10 + 5 + (5 / 5)) + 10)

= 410

Close; 4 away

Z.ai

z-ai/glm-5

z-ai

414Target
25
5
5
10
10
5

(25 + 5 + 5 + 5) * 10 + 10

= 410

Close; 4 away

openai/gpt-4o

openai

414Target
25
5
5
10
10
5

(25 + 10 + 10) * 5 + 5 * 5

= 250

Failed

Minimax

minimax/minimax-m2.5

minimax

414Target
25
5
5
10
10
5

(25+5+5+5)*10+10

= 410

Close; 4 away

anthropic/claude-sonnet-4.6

anthropic

414Target
25
5
5
10
10
5

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

= 385

Failed

MoonshotAI

moonshotai/kimi-k2.5

moonshot

414Target
25
5
5
10
10
5

((25 * 10) + (5 * 10)) + (5 * 5)

= 325

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