Open Chinese LLM Leaderboard: leaderboard
Chinese leaderboard using the official eight-task available-nonzero mean, including CLCC-H. Zero task values remain visible on task boards but are excluded from the publisher's aggregate denominator.
Metric: Official Nonzero-Task Mean (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 339 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2 72B Instruct | 75.67 | #576 |
| 2 | free-evo-qwen72B-v0.8-re | 74.62 | #435 |
| 3 | Rhea-72B-v0.5 | 74.34 | #556 |
| 4 | Smaug-72B-v0.1 | 73.16 | #611 |
| 5 | Qwen 2 72B | 72.66 | #521 |
| 6 | Le_Triomphant-ECE-TW3 | 72 | #559 |
| 7 | luxia-21.4B-alignment-v1.0 | 71.65 | #965 |
| 8 | YiSM-blossom5.1-34B-SLERP | 70.89 | #591 |
| 9 | Bagel-Hermes-34B-Slerp | 70.82 | #715 |
| 10 | UNA-SimpleSmaug-34B-v1beta | 70.7 | #712 |
| 11 | Smaug-34B-v0.1 | 70.61 | #723 |
| 12 | Yi-34Bx2-MoE-60B-DPO | 70.61 | #714 |
| 13 | Qwen 1.5 72B Chat | 70.48 | #776 |
| 14 | Luminex-34B-v0.2 | 70.44 | #696 |
| 15 | Luminex-34B-v0.1 | 70.29 | #697 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-chinese-llm-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.