The Coding Leaderboard evaluates LLMs on their software engineering capabilities. From generating production-ready code across major programming languages to designing algorithms, debugging complex functions, and translating scientific concepts into executable scripts, this category measures a model’s ability to handle a broad range of coding and software development tasks.

Scores in this category are compiled from leading coding benchmarks highlighting models that perform well on code generation, debugging, problem-solving, and technical accuracy.

Need more than code generation? Coding performance accounts for 30% of our composite rating. Explore the Overall LLM Ranking to see which top coding models also excel at autonomous agent workflows, advanced mathematics, and complex reasoning.

Last updated: November 5, 2026
Rank Model Name Relative Quality Score
1 Claude Opus 5.5
Quality of Coding 99.24%

99.24
2 Gemini 4 Carbon
Quality of Coding 96.23%

96.23
3 Claude Sonnet 5.5
Quality of Coding 94.70%

94.70
4 GPT-6 Astra
Quality of Coding 92.86%

92.86
5 Claude Fable 5.1
Quality of Coding 90.21%

90.21
6 GPT-6.1 Sol
Quality of Coding 89.86%

89.86
7 Muse Wonder Full
Quality of Coding 89.81%

89.81
8 Gemini 4 Argon
Quality of Coding 89.28%

89.28
9 ERNIE 5.0
Quality of Coding 87.97%

87.97
10 DeepSeek-V4.1-Flash
Quality of Coding 87.02%

87.02
11 Claude Fable 5
Quality of Coding 86.20%

86.20
12 Muse Wonder
Quality of Coding 86.20%

86.20
13 Kimi K3.1 Pro
Quality of Coding 86.15%

86.15
14 Muse Spark 1.3
Quality of Coding 86.15%

86.15
15 Claude Opus 5
Quality of Coding 85.08%

85.08
16 Mistral Large 4
Quality of Coding 84.96%

84.96
17 ChatGPT-4o
Quality of Coding 84.64%

84.64
18 GPT-5.5 Instant
Quality of Coding 84.06%

84.06
19 Qwen3 VL 235B A22B
Quality of Coding 84.06%

84.06
20 GPT-6.1 Terra
Quality of Coding 84.05%

84.05
21 GPT-5.6 Sol
Quality of Coding 83.84%

83.84
22 MiMo-V2.6-Pro
Quality of Coding 83.62%

83.62
23 GLM-5.3
Quality of Coding 82.86%

82.86
24 GPT-6 Sol
Quality of Coding 82.03%

82.03
25 Claude Haiku 5.5
Quality of Coding 81.94%

81.94
26 Grok 4.7
Quality of Coding 81.60%

81.60
27 Step 5
Quality of Coding 81.31%

81.31
28 Grok 4.6
Quality of Coding 80.82%

80.82
29 Kimi K3
Quality of Coding 80.28%

80.28
30 Gemini 3.8 Flash
Quality of Coding 79.57%

79.57
31 GPT-5.6 Terra
Quality of Coding 79.04%

79.04
32 Claude Opus 4.8
Quality of Coding 78.46%

78.46
33 Qwen3.8
Quality of Coding 78.23%

78.23
34 Gemini 3.7 Flash
Quality of Coding 78.15%

78.15
35 MiMo-V2.6-Flash
Quality of Coding 78.10%

78.10
36 MiniMax M3.1 Flash
Quality of Coding 78.01%

78.01
37 GLM-5.3-Flash
Quality of Coding 77.65%

77.65
38 DeepSeek-V4-Pro
Quality of Coding 77.59%

77.59
39 Hy4
Quality of Coding 77.35%

77.35
40 GPT-6 Luna
Quality of Coding 76.90%

76.90
41 Claude Opus 4.7
Quality of Coding 76.87%

76.87
42 GLM-5.2
Quality of Coding 76.56%

76.56
43 Muse Spark 1.1
Quality of Coding 76.25%

76.25
44 Qwen3.8-27B
Quality of Coding 75.57%

75.57
45 GPT-5.5
Quality of Coding 75.38%

75.38
46 Claude Sonnet 5
Quality of Coding 75.21%

75.21
47 Grok 4.5
Quality of Coding 74.90%

74.90
48 Qwen3.8-Flash-Next
Quality of Coding 74.10%

74.10
49 DeepSeek-V4-Flash
Quality of Coding 74.09%

74.09
50 Claude Opus 4.6
Quality of Coding 74.03%

74.03
51 GPT-5.6 Luna
Quality of Coding 73.59%

73.59
52 Muse Spark 1.2
Quality of Coding 73.22%

73.22
53 Qwen3.7
Quality of Coding 73.03%

73.03
54 Qwen3.7-Plus
Quality of Coding 72.93%

72.93
55 GPT-5 nano
Quality of Coding 72.75%

72.75
56 Muse Spark Open
Quality of Coding 72.52%

72.52
57 Kimi K2.6
Quality of Coding 72.50%

72.50
58 Beam 1.0
Quality of Coding 71.71%

71.71
59 Gemini 3.6 Flash
Quality of Coding 71.23%

71.23
60 Gemini 3.5 Flash
Quality of Coding 71.21%

71.21
61 Muse Spark
Quality of Coding 70.99%

70.99
62 Nemotron 3.5 Super VL
Quality of Coding 70.99%

70.99
63 Gemini 3.1 Pro
Quality of Coding 70.91%

70.91
64 MiMo-V2.5-Pro
Quality of Coding 70.42%

70.42
65 Hy3
Quality of Coding 70.32%

70.32
66 Llama-3.3 Nemotron Super 49B v1
Quality of Coding 69.42%

69.42
67 Seed 2.0 Pro
Quality of Coding 69.22%

69.22
68 Claude Sonnet 4.6
Quality of Coding 68.92%

68.92
69 Claude Opus 4.5
Quality of Coding 68.49%

68.49
70 GLM-5.1
Quality of Coding 68.47%

68.47
71 MiniMax M3
Quality of Coding 68.02%

68.02
72 GPT-5.4
Quality of Coding 67.78%

67.78
73 DeepSeek V3.1 Terminus
Quality of Coding 67.64%

67.64
74 GPT-5 mini
Quality of Coding 66.45%

66.45
75 LongCat-Flash
Quality of Coding 65.46%

65.46
76 Kimi K2
Quality of Coding 65.38%

65.38
77 Qwen3.6 Plus
Quality of Coding 65.34%

65.34
78 MiMo-V2-Omni
Quality of Coding 64.85%

64.85
79 Llama 3.1 Nemotron 70B
Quality of Coding 64.49%

64.49
80 Inkling-Small
Quality of Coding 64.30%

64.30
81 Step-3.5-Flash
Quality of Coding 64.25%

64.25
82 Claude Opus 4
Quality of Coding 63.96%

63.96
83 MiMo-V2.5
Quality of Coding 63.93%

63.93
84 Inkling
Quality of Coding 63.59%

63.59
85 MiniMax M2.7
Quality of Coding 63.43%

63.43
86 Gemini 3 Pro
Quality of Coding 63.31%

63.31
87 GLM-5
Quality of Coding 63.23%

63.23
88 Granite 4.2 30B
Quality of Coding 63.22%

63.22
89 Kimi K2.5
Quality of Coding 62.78%

62.78
90 MiMo-V2-Pro
Quality of Coding 62.71%

62.71
91 Kimi K2.7 Code
Quality of Coding 62.46%

62.46
92 Grok 4 Fast
Quality of Coding 62.25%

62.25
93 Jamba 1.5 Large
Quality of Coding 62.17%

62.17
94 Mistral Medium 3.1
Quality of Coding 62.14%

62.14
95 GPT-5.4 nano
Quality of Coding 62.02%

62.02
96 Claude Sonnet 4
Quality of Coding 61.89%

61.89
97 Qwen3
Quality of Coding 61.83%

61.83
98 GPT OSS 20B
Quality of Coding 61.81%

61.81
99 GPT-5.4 mini
Quality of Coding 61.75%

61.75
100 GPT-5
Quality of Coding 61.69%

61.69