Skip to main content

Reasoning Text Model Leaderboard

Expert problems, research mathematics, logic, games, and pattern discovery.

Sources
1Nov 9, 2026
100%22 items
84.4
2Sep 3, 2026
100%27 items
80.4
3Sep 22, 2026
100%27 items
76.9
4Oct 8, 2026
100%27 items
76.8
5Oct 26, 2026
63%18 items
75.3
6Sep 28, 2026
100%27 items
73.8
7Sep 30, 2026
100%27 items
72.0
8Sep 1, 2026
100%26 items
70.8
9Apr 23, 2026
63%20 items
70.6
10Jun 9, 2026
75%22 items
66.3
11Sep 22, 2026
100%25 items
65.2
12Jul 24, 2026
100%26 items
65.0
13Jul 9, 2026
100%25 items
64.9
14Oct 17, 2026
75%21 items
60.9
15Oct 19, 2026
38%15 items
60.8
16May 28, 2026
75%22 items
59.9
17Jul 9, 2026
88%24 items
59.8
18Apr 8, 2026
38%16 items
59.6
19Aug 5, 2026
38%18 items
58.7
20Sep 2, 2026
100%26 items
57.9
21Oct 7, 2026
100%27 items
57.7
22Jul 8, 2026
50%20 items
56.6
23Apr 17, 2026
25%15 items
56.0
24Aug 3, 2026
88%23 items
56.0
25Sep 2, 2026
100%24 items
55.9
26Jun 30, 2026
75%22 items
55.5
27Aug 13, 2026
88%23 items
54.4
28Jul 16, 2026
88%26 items
54.1
29May 21, 2026
63%17 items
53.9
30Jun 13, 2026
63%20 items
52.9
31Aug 12, 2026
100%26 items
52.8
32Apr 22, 2026
25%15 items
52.4
33Jul 31, 2026
75%18 items
52.1
34Jul 21, 2026
75%20 items
52.1
35Oct 5, 2026
100%27 items
51.6
36Sep 21, 2026
63%21 items
51.6
37Sep 10, 2026
100%25 items
51.4
38Sep 21, 2026
100%26 items
51.3
39Jul 9, 2026
100%25 items
49.7
40Sep 20, 2026
63%19 items
49.7
41Sep 22, 2026
100%27 items
48.9
42Apr 22, 2026
25%12 items
48.4
43Jun 12, 2026
50%15 items
48.1
44Apr 21, 2026
63%18 items
47.8
45Mar 18, 2026
25%15 items
47.2
46Jun 2, 2026
50%14 items
46.9
47Aug 26, 2026
88%24 items
46.3
48Mar 17, 2026
63%20 items
45.4
49Jul 23, 2026
25%11 items
45.4
50Feb 17, 2026
50%19 items
45.0
51Jul 30, 2026
63%21 items
44.3
52Apr 7, 2026
50%17 items
43.6
53Jul 15, 2026
63%23 items
42.7
54Jun 1, 2026
63%21 items
42.6
55Jul 21, 2026
75%21 items
38.4
56Oct 15, 2025
50%19 items
35.3

Not yet ranked

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

78.7Fewer than 3 coding evaluation types

Frequently asked questions