Released: 2026-06-18
openai/gpt-4o
openai
5
6
1
8
3
9
4
7
2
7
3
9
4
1
8
2
6
5
2
9
8
5
6
7
3
4
1
4
5
2
6
7
3
9
1
8
6
8
7
1
9
5
5
2
3
3
1
6
2
8
4
7
5
6
9
4
3
2
5
6
1
3
7
1
7
4
7
2
9
6
8
9
8
2
5
3
4
4
5
9
7
google/gemini-3-flash-preview
7
5
8
9
2
1
4
6
3
6
1
3
4
9
8
2
7
5
2
9
4
5
6
7
1
8
3
5
2
9
6
4
3
7
1
8
8
6
7
1
5
2
3
9
4
4
3
1
8
7
9
5
2
6
9
7
6
2
1
5
8
3
4
1
4
2
7
3
6
9
5
8
3
8
5
9
8
4
6
2
7
deepseek/deepseek-chat-v3-0324
deepseek
5
7
6
9
1
2
4
8
3
3
1
4
5
8
6
2
7
9
2
9
8
3
4
7
6
5
1
4
3
2
6
9
5
7
1
8
6
5
7
8
2
1
3
9
4
8
2
9
4
7
3
5
6
1
9
4
1
2
5
8
7
3
6
1
6
3
7
9
4
8
2
5
7
8
5
1
6
3
9
4
2
mistralai/mistral-medium-3
mistral
8
3
6
9
4
5
4
7
1
4
7
1
3
2
8
2
9
5
2
9
5
4
6
7
8
1
3
5
2
4
6
3
9
7
1
8
6
1
7
1
8
2
3
5
9
3
8
9
5
1
7
6
4
2
9
4
8
2
5
1
6
3
7
1
6
2
7
9
3
5
8
4
7
5
3
8
4
6
1
2
9
meta-llama/llama-3.3-70b-instruct
meta
5
3
4
6
7
8
9
1
2
6
7
2
1
9
5
3
4
8
1
9
8
3
4
2
5
6
7
8
5
9
7
6
1
4
2
3
4
2
6
8
5
3
7
9
1
7
1
3
9
2
4
8
5
6
9
6
1
5
3
7
2
8
4
2
8
7
4
1
9
6
3
5
3
4
5
2
8
6
1
7
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.