LiveCodeBench
coding.
v6coding
Definition
- Version
- v6
- Category
- coding
- Maintainer
- not in the notes
- Status
- not in the notes
- Home page
- not in the notes
Sources
- Qwen/Qwen3.8-Flash-Next , Hugging Face huggingface.co https://huggingface.co/Qwen/Qwen3.8-Flash-Next
Scores at a Glance
Score in %
Higher is better for this unit.
1
Qwen3.8-Flash-NextAlibaba (Qwen)
Qwen3.8-27BAlibaba (Qwen)
Qwen3.8-Flash-NextAlibaba (Qwen)91.9 %
vendor-reported2
Qwen3.8-27BAlibaba (Qwen)90.3 %
vendor-reported0255075100
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
Qwen3.8-Flash-Next | 91.9 % | vendor-reported | – | – | – | huggingface.co |
Qwen3.8-27B | 90.3 % | vendor-reported | – | – | – | huggingface.co |
A.X K2 | 84 % | vendor-reported | Thinking mode; Feb to May window | A.X K2 model card | Jul 2026 | huggingface.co |
MiMo-V2-Flash | 80.6 % | vendor-reported | – | MiMo-V2-Flash model card, post-trained model table | 16 Dec 2025 | huggingface.co |
EXAONE 4.0 (32B) | 66.7 % | vendor-reported | Reasoning mode | EXAONE 4.0 model card | Jul 2025 | huggingface.co |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "LiveCodeBench", https://aurion.semai.id/benchmarks/livecodebench-v6/, records of October 2026.


