AI Coding Models Leaderboard

Multi-source blend of coding benchmarks into a single AISOTA score.

Data as of 2026-09-2850 models
#ModelArtificial Analysis AgenticLiveBench CodingAISOTA Score
1SOTAdeepseek-v4.1-flash-max—98.498.4
2claude-opus-5-5-xhigh-effort—96.896.8
3claude-fable-5-1-max-effort—95.295.2
4smaug-agentic—93.593.5
5claude-fable-5-max-effort—91.991.9
6claude-opus-5-max-effort—90.390.3
7muse-spark-1.3-xhigh—88.788.7
8kimi-k3—87.187.1
9glm-5.3—85.585.5
10qwen3.8-max—83.983.9
11claude-sonnet-5-xhigh-effort—82.382.3
12smaug-flash—80.680.6
13gpt-5.6-sol-max—79.079.0
14qwen3.8-27b—77.477.4
15smaug-mini—75.875.8
16gpt-6-astra-max—74.274.2
17gemini-3.7-flash-high—72.672.6
18deepseek-v4-flash-vision-exp—71.071.0
19qwen3.8-flash-next—69.469.4
20muse-spark-1.1-xhigh—67.767.7
21union-alpha—66.166.1
22glm-5.3-flash—64.564.5
23muse-spark-1.2-xhigh—62.962.9
24gpt-5.5-xhigh—61.361.3
25grok-4.6—59.759.7
26gpt-6-sol-max—58.158.1
27gpt-5.6-terra-max—56.556.5
28deepseek-v4-pro-0813—54.854.8
29gpt-5.4-xhigh—53.253.2
30grok-4.7-xhigh—51.651.6
31claude-opus-4-7-xhigh-effort—50.050.0
32gpt-5.2-codex—48.448.4
33claude-opus-4-8-max-effort—46.846.8
34glm-5.2—45.245.2
35gpt-6-luna-max—43.543.5
36gpt-5.6-luna-max—41.941.9
37ox-alpha-max—40.340.3
38gemini-3.8-flash-high—38.738.7
39grok-4.5—37.137.1
40claude-opus-4-6-thinking-auto-high-effort—35.535.5
41gemini-3.5-flash-high—33.933.9
42gpt-5.2-2025-12-11-high—32.332.3
43kimi-k2.6-thinking—30.630.6
44deepseek-v4-flash-0731—29.029.0
45inkling-xhigh—27.427.4
46gemini-3.5-flash-lite-high—25.825.8
47claude-sonnet-4-6-thinking-auto-medium-effort—24.224.2
48gemini-3.6-flash-high—22.622.6
49gemini-3.1-pro-preview-high—21.021.0
50kimi-k2.7-code—19.419.4