Benchgen

SWE-bench Verified — Results

RankModelScore
1claude-fable-595
2claude-mythos-preview93.9
3claude-opus-4-888.6
4claude-opus-4-787.6
5claude-sonnet-585.2
6claude-opus-4-681.42
7claude-opus-4-580.9
8deepseek-v4-pro-max80.6
9gemini-3-1-pro80.6
10minimax-m380.5
11qwen3-7-max80.4
12qwen3-7-max80.4
13kimi-k2-680.2
14minimax-m2-580.2
15kimi-k2-090580.2
16gpt-5-280
17claude-sonnet-4-679.6
18claude-sonnet-4-679.6
19deepseek-v4-flash-max79
20mimo-v2-5-pro78.9
21qwen3-6-plus78.8
22gemini-3-flash78
23mimo-v2-pro78
24hy378
25glm-577.8

SWE-bench Verified — Phases

1 phaseActive

Phase 1·

SWE-bench Verified

Jun 19, 2026
Current