# Leaderboard methodology · 88z.ai

Last collected by this site: 2026-11-15 06:40 (UTC+8, Asia/Shanghai). Source publication dates are in the table or the citations below.

- Artificial Analysis Intelligence Index: 28%
- LMArena Text Arena: 28%
- LiveBench 2.0 overall: 22%
- Coding composite: 15%
- Reasoning composite: 7%

Sources:
- [LMArena Text Arena (style-controlled)](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)
- [LMArena WebDev Arena](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)
- [LMArena Vision Arena (style-controlled)](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)
- [LMArena Text-to-Image Arena](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)
- [LMArena Text-to-Video Arena](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)
- [LMArena Agent Arena](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)
- [Artificial Analysis Intelligence Index](https://artificialanalysis.ai/leaderboards/models)
- [LiveBench 2.0](https://livebench.ai/)
- [Aider Polyglot](https://aider.chat/docs/leaderboards/)
- [SWE-bench Verified (mini-SWE-agent)](https://www.swebench.com/verified)
- [SuperCLUE Chinese general board](https://www.superclueai.com/generalpage)