Back to home
Coding
How well can each AI model write, debug, and refactor code? Ranked by Artificial Analysis Coding Index.
RankModelScore
1
23
4567891011
1213
141516
171819202122
23242526
27282930
GPT-5openai/gpt-5-6-sol-xhigh
78.3
GPT-5openai/gpt-5-6-terra
76.7
Claude Fable 5anthropic/claude-fable-5
76.5
K
Kimi K3kimi/kimi-k3
76.2
GPT-5openai/gpt-5-5
74.9
Claude Opus 4.8anthropic/claude-opus-4-8
74.3
Claude Opus 4.7anthropic/claude-opus-4-7
73.6
S
Grok 4xai/grok-4-5
72.4
Claude Sonnet 5anthropic/claude-sonnet-5
71.5
GPT-5openai/gpt-5-6-luna
71.4
Muse Sparkmeta/muse-spark-1-1
71.3
GPT-5openai/gpt-5-4
71.1
Gemini 3.5 Flashgoogle/gemini-3-5-flash
70.1
Gemini 3.1 Progoogle/gemini-3-1-pro-preview
68.8
Z
GLM 5.2zai/glm-5-2
68.8
Qwen3.7 Maxalibaba/qwen3-7-max
66.0
Claude Sonnet 4.6anthropic/claude-sonnet-4-6-adaptive
63.0
K
Kimi K2.6kimi/kimi-k2-6
61.8
K
Kimi K2.7 Codekimi/kimi-k2-7-code
60.8
X
MiMo V2.5 Proxiaomi/mimo-v2-5-pro
60.2
K
KAT Coder Pro V2kwaikat/kat-coder-pro-v2
59.5
DeepSeek V4 Prodeepseek/deepseek-v4-pro
59.4
N
Nex N2 Pronex/nex-n2-pro
59.1
T
Hy3tencent/hy3
58.8
Muse Sparkmeta/muse-spark
58.6
MiniMax M3minimax/minimax-m3
58.6
X
MiMo V2.5xiaomi/mimo-v2-5-0424
56.8
DeepSeek V4 Flashdeepseek/deepseek-v4-flash
56.2
GPT-5openai/gpt-5-4-nano
56.1
Qwen3.7 Plusalibaba/qwen3-7-plus
55.9
135 models tested