LiveCodeBench v6
6 models tested · Updated 2026-04-20 · Verified sources only
Kimi K2.6 leads at 89.6%
1
Moonshot AI · HuggingFace/moonshotai · 2026-04-20
Near frontier-level coding. Trails Gemini 3.1 Pro (91.7) by 2 points.
89.6%
2
Alibaba · HuggingFace/Qwen · 2026-04-22
Near-frontier coding benchmark for sub-30B.
83.9%
3
Xiaomi · HuggingFace/XiaomiMiMo-MiMo-V2-Flash · 2026-01-06
LiveCodeBench v6. Strong coding.
80.6%
4
OpenAI · arxiv/2604.08644 · 2026-04-14
From EXAONE 4.5 technical report. Trails EXAONE 4.5 33B (81.4).
78.1%
5
Google DeepMind · HuggingFace/google-gemma-4-12b-it · 2026-05-23
LiveCodeBench v6. Strong coding for 12B dense.
72.0%
6
Tencent · GitHub/Tencent-Hunyuan-Hy3-preview · 2026-04-23
Beats Kimi-K2 base (30.86) and GLM-4.5 base (27.43).
34.86%