Released: 2026-08-05
openai/gpt-4o
openai
4
9
1
6
2
8
3
5
7
6
5
8
7
4
9
2
3
1
7
2
3
1
5
3
9
4
6
1
3
4
5
7
2
6
9
8
9
6
7
4
8
3
5
2
1
2
8
5
9
1
6
4
7
3
8
7
2
3
1
5
6
1
4
5
4
6
2
9
7
1
8
9
9
1
3
8
6
4
7
6
2
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
google/gemini-3-flash-preview
4
5
1
6
7
2
3
8
9
3
6
8
5
4
9
2
7
1
7
2
9
8
3
1
5
4
6
1
7
4
2
5
3
9
6
8
6
9
7
4
8
0
5
2
3
2
8
5
1
9
6
4
7
3
8
4
3
7
1
5
6
9
2
5
3
2
9
6
7
8
1
4
9
1
6
3
2
4
7
5
8
mistralai/mistral-medium-3
mistral
1
4
5
6
8
2
3
9
7
3
6
8
7
5
9
4
2
1
7
2
9
1
3
4
8
4
6
6
5
4
2
9
3
7
1
8
1
8
7
4
6
1
5
2
9
2
9
3
5
7
6
1
7
4
8
7
2
9
1
5
6
3
4
5
3
6
8
4
7
9
5
2
9
1
4
3
2
8
7
6
5
deepseek/deepseek-chat-v3-0324
deepseek
1
9
5
6
4
2
3
8
7
4
7
3
8
5
9
2
6
1
7
2
6
1
3
5
8
4
9
6
3
4
2
9
1
7
5
8
8
6
7
4
5
3
5
2
6
2
5
1
9
8
6
4
7
3
8
4
2
5
1
8
6
9
4
5
8
9
4
2
7
1
3
2
9
1
8
3
6
4
9
8
5
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.