Atlas

Benchmarks

← All benchmarks

Artificial Analysis Coding Index

Indexes · 2025-02-13

Composite index of coding and software engineering benchmarks.

Top models (higher is better)

ModelScore
GPT-5.6 Sol78.3
GPT-5.6 Terra76.7
Claude Fable 576.5
Kimi K376.2
GPT-5.574.9
Opus 4.874.3
Opus 4.773.6
Grok 4.572.4
Sonnet 571.5
GPT-5.6 Luna71.4
Muse Spark 1.171.3
GPT-5.471.1
Gemini 3.5 Flash70.1
Gemini 3.6 Flash69.2
DeepSeek-V4-Flash-073169.1
Gemini 3.1 Pro Preview68.8
GLM-5.268.8
Qwen3.7-Max66.0
Sonnet 4.663.0
Motif-3-Beta62.0
Kimi K2.661.8
Kimi K2.7 Code60.8
MiMo-V2.5-Pro60.2
KAT-Coder-Pro V259.5
DeepSeek-V4-Pro59.4
Nex-N2-Pro59.1
Hy358.8
Agnes 2.5 Pro Alpha58.8
Muse Spark58.6
MiniMax M358.6
MiMo-V2.556.8
DeepSeek-V4-Flash56.2
GPT-5.4 Mini56.1
GPT-5.4 Nano56.1
Qwen3.7-Plus55.9
GLM-5.155.8
Qwen3.6 Plus (2026-04-02)54.5
Qwen3.6 27B53.7
Inkling-Small52.9
MiniMax M2.752.6
Loading Atlas data…