AI vs Puzzles

Back to Home
Puzzle#122

Countdown

Released: 2026-02-26


10
Models Tested
4
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

mistralai/mistral-medium-3

mistral

267Target
75
4
6
8
4
10

(75 * (8 + 4)) + (10 * (6 + 4))

= 1000

Failed

google/gemini-2.0-flash-001

google

267Target
75
4
6
8
4
10

(75 * 4) - 6 - 4 - 10 + 8

= 288

Failed

openai/gpt-4o

openai

267Target
75
4
6
8
4
10

(75 * 4) - (6 + 8 + 4)

= 282

Failed

google/gemini-3-flash-preview

google

267Target
75
4
6
8
4
10

(4 * (75 - 8)) - (10 - 6)

= 264

Close; 3 away

deepseek/deepseek-chat-v3-0324

deepseek

267Target
75
4
6
8
4
10

(75 * 4) - (10 + 8 + 6 + 4)

= 272

Close; 5 away

anthropic/claude-3.7-sonnet

anthropic

267Target
75
4
6
8
4
10

(75 - 4) * 4 - 6 + 10

= 288

Failed

Z.ai

z-ai/glm-5

z-ai

267Target
75
4
6
8
4
10

75 + (8 * 6 * 4)

= 267

Perfect Solution!

Minimax

minimax/minimax-m2.5

minimax

267Target
75
4
6
8
4
10

(8*6*4)+75

= 267

Perfect Solution!

MoonshotAI

moonshotai/kimi-k2.5

moonshot

267Target
75
4
6
8
4
10

(75 - 8) * 4 - (10 - 4) / 6

= 267

Perfect Solution!

x-ai/grok-3-mini-beta

x-ai

267Target
75
4
6
8
4
10

75 + (8 * 6 * 4)

= 267

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