the ai forger
Models
Organizations
Providers
Benchmarks
Compare

SWE-Bench Verified

general
text
About

SWE-Bench Verified benchmark

Evaluation Stats
Total Models28
Organizations5
Verified Results0
Self-Reported28
Benchmark Details
Max Score1
Language
en
Performance Overview
Score distribution and top performers

Score Distribution

28 models
Top Score
74.9%
Average Score
48.5%
High Performers (80%+)
0

Top Organizations

#1Mistral AI
2 models
57.6%
#2Anthropic
6 models
54.9%
#3Google
5 models
49.1%
#4OpenAI
11 models
45.6%
#5DeepSeek
4 models
41.4%
Leaderboard
Top 20 models ranked by performance
1
74.9%
Raw: 0.749
Self-reported
2
72.7%
Raw: 0.727
Self-reported
3
72.5%
Raw: 0.725
Self-reported
4
70.3%
Raw: 0.703
Self-reported
5
69.1%
Raw: 0.691
Self-reported
6
68.1%
Raw: 0.681
Self-reported
7
67.2%
Raw: 0.672
Self-reported
8
63.2%
Raw: 0.632
Self-reported
9
61.6%
Raw: 0.616
Self-reported
10
60.4%
Raw: 0.604
Self-reported
11
57.6%
Raw: 0.576
Self-reported
12
54.6%
Raw: 0.546
Self-reported
13
53.6%
Raw: 0.536
Self-reported
14
49.3%
Raw: 0.493
Self-reported
15
49.2%
Raw: 0.492
Self-reported
16
49.0%
Raw: 0.49
Self-reported
17
42.0%
Raw: 0.42
Self-reported
18
41.3%
Raw: 0.413
Self-reported
19
41.0%
Raw: 0.41
Self-reported
20
40.6%
Raw: 0.406
Self-reported
Showing top 20 of 28 models