Inference
EN
Artificial Analysis

Artificial Analysis LiveCodeBench

An independently evaluated LiveCodeBench result from Artificial Analysis.

ModelsModels with a result.12
Top resultBest result on this benchmark.91.7%Gemini 3 Pro
Top-3 spreadPoints from first to third.3.9 pts
YearYear of release.2026

Result and price

8 of 12 results

Results

12 results
#ModelResultThe score from the source.
01Gemini 3 Pro91.7%
02GLM-4.789.4%
03GPT-OSS 120BAbove fitFar above this model's index.87.8%
04Claude Opus 4.5 Thinking87.1%
05GPT-5.186.8%
06o4-mini85.9%
07GPT-5 (high)84.6%
08Grok 4 Fast (Reasoning)83.2%
09Grok 4.1 Fast (Reasoning)82.2%
10Grok 481.9%
11o380.8%
12Gemini 2.5 Pro80.1%