Released: 2026-07-07
openai/gpt-4o
openai
7
1
6
2
5
9
4
3
8
3
4
5
8
9
7
2
1
6
9
8
2
3
1
6
5
4
7
4
5
7
1
6
2
8
9
3
2
6
1
7
8
3
9
4
5
5
3
9
4
7
1
6
2
4
8
9
4
5
6
2
3
6
1
1
7
3
9
4
8
7
8
2
6
2
8
7
3
4
1
5
9
google/gemini-3-flash-preview
7
1
6
2
5
8
4
9
3
3
4
5
1
9
7
2
8
6
9
2
8
3
1
6
5
4
7
4
5
7
9
6
2
8
3
1
2
6
1
8
4
3
9
7
5
5
3
9
7
8
1
6
2
4
8
9
4
5
7
1
3
6
2
1
7
5
6
2
9
3
4
8
6
8
2
4
3
1
7
5
9
meta-llama/llama-3.3-70b-instruct
meta
5
1
9
7
5
8
4
3
2
3
4
7
2
1
7
2
9
6
2
8
6
5
4
1
7
1
7
4
5
3
9
6
2
1
8
7
1
2
5
8
9
3
9
6
5
9
7
1
4
7
5
8
2
3
8
6
4
1
2
9
5
7
9
7
3
2
6
8
4
6
5
1
6
9
8
3
7
5
3
4
9
mistralai/mistral-medium-3
mistral
2
1
6
9
5
4
3
8
7
3
4
5
8
3
7
2
1
6
9
8
7
2
1
6
4
3
5
4
5
3
7
6
8
1
9
2
7
6
1
4
2
3
9
5
8
8
3
9
5
9
1
6
2
4
8
2
4
3
7
5
8
6
1
1
7
3
6
8
2
5
4
9
6
9
8
1
4
9
7
5
3
deepseek/deepseek-chat-v3-0324
deepseek
7
1
6
8
5
2
4
3
9
3
9
5
4
8
7
2
1
6
2
8
4
3
1
6
5
9
7
4
5
1
9
6
8
7
2
3
8
2
7
6
4
3
9
1
5
6
3
9
5
7
1
8
2
4
8
4
2
1
9
5
3
6
7
5
7
3
2
6
9
1
4
8
6
9
1
7
3
4
2
5
9
Each model receives the same Sudoku puzzle with a partially filled grid. Models are tasked with solving the puzzle by filling in all empty cells following Sudoku rules, and their solutions are evaluated without feedback or retries.