Open Chinese LLM - ARC Challenge — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 338 models tracked.

Top models

#ModelScore
1free-evo-qwen72B-v0.8-re71.16
2TW3-JRGL-v271.16
3Rhea-72B-v0.570.99
4luxia-21.4B-alignment-v1.069.97
5MultiVerse_70B69.28
6Le_Triomphant-ECE-TW369.2
7Smaug-72B-v0.167.66
8Qwen 2 72B Instruct66.47
9UNA-SimpleSmaug-34B-v1beta66.04
10Smaug-34B-v0.165.87
11luxia-21.4B-alignment-v1.264.25
12ECE-TW3-JRGL-V564.16
13Yi-34Bx2-MoE-60B-DPO64.08
14Mixtral_34Bx2_MoE_60B63.99
15Qwen 2 72B63.74

Interactive version: theaggregate.ai/benchmark?slug=open-chinese-llm-arc-challenge · How the rankings work · Data refreshed daily, snapshot 2026-07-22.