Leaderboard · Nov 2026

Best AI models for coding, ranked by developers

The main table covers models with at least 100 opinions. Less-discussed models are listed separately, because a high score from a few dozen people is a weaker signal.

Widely discussed models

Ranked by approval, adjusted for sample size. Model pages focus on coding use, not general chat.

All 13 models
Ranked by approval, adjusted for sample size
#ModelApprovalLast 90 daysOpinions
01DeepSeek V4 Flash
83%±5
80%
151
02Claude Opus 4.5
78%±3.8
90%
322
03Claude Opus 5.5
75%±5.2
75%
186
04GLM 5.2
75%±6
82%
142
05DeepSeek V4 Pro
75%±6.5
67%
121
06GPT-5.4
72%±6.9
–
114
07GPT-5.5
69%±5
67%
236
08GPT-5.6 Sol
67%±4.9
66%
247
09Claude Opus 4.6
63%±4.7
63%
287
10Claude Fable 5
57%±5.9
48%
188
11Claude Opus 4.8
49%±7.1
57%
134
12Claude Opus 4.7
42%±6.7
–
145
13Claude Opus 5
33%±6.1
30%
162

Shown: models with at least 100 opinions. Approval is the positive share of opinions that took a side; rank uses the lower bound of a 90% confidence interval.

Less-discussed models

Fewer than 100 opinions each. Promising, but treat the scores as early.

All 32 models
Ranked by approval, adjusted for sample size
#ModelApprovalLast 90 daysOpinions
14GLM 5.3
85%±7.2
85%
67
15Qwen 3 Coder
87%±8.9
–
39
16Gemini 3 Flash
85%±10.5
–
31
17DeepSeek V4.1 Flash
80%±8.5
80%
60
18GLM 5.1
78%±8.3
–
67
19Qwen 3.6
76%±7.2
71%
94
20Qwen 3.6 27B
76%±8.1
56%
76
21DeepSeek V4
78%±9.8
72%
48
22Kimi K3
73%±7.5
74%
94
23GLM 5
78%±12.4
–
30
24GPT-5.3 Codex
74%±9.8
–
54
25Kimi K2.5
74%±11.3
–
41
26GPT-5.6
76%±12.2
75%
33
27Qwen 3 Coder Next
72%±10.7
–
48
28Qwen 3.8 27B
69%±9.1
69%
70
29GLM 5.3 Flash
71%±11.5
71%
42
30GPT-5.2
68%±8.5
–
82
31Qwen 3.6 35B A3B
70%±11.9
56%
40
32Kimi K2.6
70%±12.2
–
38
33Claude Fable 5.1
69%±12.9
69%
35
34Claude Sonnet 4.6
65%±9.7
–
65
35Gemma 4
62%±9.7
–
68
36Claude Sonnet 4.5
57%±8.3
–
97
37Qwen 3.5
61%±12.5
–
41
38Claude Sonnet 5
61%±13
41%
38
39GPT-5
57%±10.3
–
62
40Gemini 3
56%±9.8
–
70
41GLM 4.7
56%±12.5
–
43
42Gemini 3.1 Pro
54%±13.7
–
36
43Gemini 3 Pro
52%±12.7
–
42
44GPT-6 Astra
51%±11.9
51%
48
45Qwen 3.8
50%±11.3
50%
53

Small samples: read the opinion count before the percentage.