Skip to main content
Meo

LiveCodeBench — AI model leaderboard

AI models ranked by LiveCodeBench, an aggregated third-party benchmark from artificial_analysis. Higher is better. Cross-referenced against our first-party meo scores and Effective Value (𝕍).

Ranking 38 models across the full field · as of 2026-06-07.

#ModelLabLiveCodeBench
1OpenAI: GPT-5.2 Chatopenai88.9%
2OpenAI: gpt-oss-120bopenai87.8%
3OpenAI: GPT-5.1openai86.8%
4OpenAI: o4 Miniopenai85.9%
5OpenAI: GPT-5.1-Codexopenai84.9%
6OpenAI: GPT-5openai84.6%
7OpenAI: GPT-5 Codexopenai84.0%
8OpenAI: GPT-5 Miniopenai83.8%
9OpenAI: GPT-5.1-Codex-Miniopenai83.6%
10OpenAI: o3openai80.8%
11Google: Gemini 2.5 Progoogle80.1%
12Google: Gemini 3 Flash Previewgoogle79.7%
13OpenAI: GPT-5 Nanoopenai78.9%
14OpenAI: gpt-oss-20bopenai77.7%
15Prime Intellect: INTELLECT-3prime-intellect77.7%
16OpenAI: o3 Mini Highopenai73.4%
17OpenAI: o3 Miniopenai71.7%
18OpenAI: o1openai67.9%
19Google: Gemini 2.5 Flash Lite Preview 09-2025google64.1%
20Google: Gemini 2.5 Flashgoogle49.5%
21OpenAI: GPT-4.1 Miniopenai48.3%
22OpenAI: GPT-4.1openai45.7%
23Reka Flash 3rekaai43.5%
24Xiaomi: MiMo-V2-Flashxiaomi40.2%
25Meta: Llama 4 Maverickmeta-llama39.7%
26OpenAI: GPT-4o (2024-05-13)openai33.4%
27OpenAI: GPT-4.1 Nanoopenai32.6%
28OpenAI: GPT-4o (2024-08-06)openai31.7%
29OpenAI: GPT-4oopenai30.9%
30Meta: Llama 4 Scoutmeta-llama29.9%
31OpenAI: GPT-4 Turboopenai29.1%
32Cohere: Command Acohere28.7%
33OpenAI: GPT-4o-miniopenai23.4%
34Microsoft: Phi 4microsoft23.1%
35Google: Gemma 3 27Bgoogle13.7%
36Google: Gemma 3 12Bgoogle13.7%
37Microsoft: Phi 4 Mini Instructmicrosoft12.6%
38Google: Gemma 3 4Bgoogle11.2%

Artificial Analysis (artificialanalysis.ai). Redistribution requires an AA commercial license.

← All rankingsMethodology & 𝕍 →