Released: 2026-06-23
openai/gpt-4o
openai
2
6
5
7
1
4
9
3
8
3
4
7
5
8
9
6
1
2
8
1
9
6
3
2
5
4
7
6
2
8
3
7
5
4
8
9
9
7
1
2
6
3
8
5
5
4
3
5
1
9
8
7
6
2
1
9
6
4
5
7
3
2
6
7
8
2
9
4
6
1
7
3
5
3
4
8
2
1
2
9
6
google/gemini-3-flash-preview
6
2
5
7
3
4
9
8
1
3
9
7
5
8
1
6
4
2
8
1
4
6
9
2
5
3
7
2
7
8
4
6
5
1
3
9
9
6
1
2
7
3
8
4
5
4
5
3
1
9
8
7
6
2
1
4
6
8
5
7
3
2
9
2
8
9
9
4
6
2
5
3
5
3
3
3
2
9
4
1
6
mistralai/mistral-medium-3
mistral
8
6
5
3
7
4
9
1
2
3
9
7
8
1
2
6
4
5
2
1
4
6
9
5
8
3
7
7
2
6
1
3
5
4
8
9
9
3
1
4
6
8
2
7
5
4
5
8
2
9
7
1
6
3
6
4
2
7
5
9
3
8
1
5
8
9
3
4
6
7
2
1
1
7
3
5
8
1
5
9
6
meta-llama/llama-3.3-70b-instruct
meta
5
3
4
8
1
6
9
2
7
3
9
7
5
4
8
1
6
2
1
2
6
3
7
9
5
8
4
9
5
8
2
3
1
7
4
6
7
4
1
9
6
5
8
3
2
2
8
3
4
5
7
6
1
9
6
1
5
7
2
3
4
9
8
8
6
9
1
4
2
3
7
5
4
7
2
6
8
5
1
3
9
deepseek/deepseek-chat-v3-0324
deepseek
2
6
5
3
7
4
9
8
1
3
4
7
5
8
9
6
2
1
8
1
9
6
1
2
4
5
7
6
2
8
1
4
5
7
3
9
7
3
1
2
9
6
8
4
5
4
5
9
8
3
7
2
6
1
1
9
6
4
5
8
3
7
2
5
8
2
9
7
6
1
4
3
9
7
3
1
2
5
4
8
6
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.