Skip to main content

Coding Text Model Leaderboard

Terminal tasks, software engineering, scientific computing, migration, and reconstruction.

Sources
1Nov 9, 2026
80%22 items
85.0
2Sep 22, 2026
100%27 items
82.7
3Sep 28, 2026
100%27 items
75.9
4Sep 1, 2026
100%26 items
72.2
5Sep 3, 2026
100%27 items
72.2
6Sep 30, 2026
100%27 items
70.7
7Oct 26, 2026
60%18 items
68.6
8Oct 8, 2026
100%27 items
68.4
9Jul 24, 2026
100%26 items
66.0
10Sep 22, 2026
100%25 items
63.4
11Jun 9, 2026
100%22 items
63.3
12Jul 9, 2026
100%25 items
60.7
13Oct 5, 2026
100%27 items
60.1
14Oct 17, 2026
80%21 items
58.9
15Sep 20, 2026
80%19 items
58.8
16Sep 2, 2026
100%24 items
58.2
17Oct 7, 2026
100%27 items
58.1
18Sep 21, 2026
100%21 items
57.4
19Jul 9, 2026
100%24 items
55.3
20Sep 21, 2026
100%26 items
55.1
21Oct 19, 2026
60%15 items
54.2
22Apr 23, 2026
80%20 items
54.0
23May 28, 2026
100%22 items
53.5
24Aug 12, 2026
100%26 items
52.7
25Sep 10, 2026
100%25 items
52.3
26Sep 2, 2026
100%26 items
51.6
27Aug 26, 2026
100%24 items
51.5
28Apr 8, 2026
80%16 items
50.0
29Aug 3, 2026
100%23 items
49.9
30Sep 22, 2026
100%27 items
49.6
31Jun 30, 2026
100%22 items
49.1
32Jul 9, 2026
100%25 items
48.8
33Feb 17, 2026
80%19 items
48.7
34Jul 16, 2026
100%26 items
48.5
35Aug 13, 2026
100%23 items
48.3
36Jul 21, 2026
80%20 items
48.3
37Jun 13, 2026
80%20 items
48.3
38Jul 31, 2026
100%18 items
48.1
39Jul 8, 2026
100%20 items
47.3
40Aug 5, 2026
100%18 items
46.4
41Apr 21, 2026
80%18 items
45.8
42Jun 12, 2026
80%15 items
44.1
43Apr 7, 2026
80%17 items
43.3
44May 21, 2026
60%17 items
42.0
45Mar 18, 2026
80%15 items
41.4
46Mar 17, 2026
100%20 items
40.4
47Apr 17, 2026
80%15 items
40.4
48Jun 2, 2026
60%14 items
40.2
49Apr 22, 2026
80%15 items
40.2
50Jul 30, 2026
100%21 items
39.8
51Apr 22, 2026
60%12 items
39.3
52Jun 1, 2026
80%21 items
39.1
53Oct 15, 2025
80%19 items
38.4
54Jul 15, 2026
100%23 items
38.4
55Jul 21, 2026
80%21 items
38.0
56Jul 23, 2026
60%11 items
34.7

Not yet ranked

These models have partial scores and will receive a rank when evidence is complete.

81.4Fewer than 3 coding evaluation types

Frequently asked questions