SeaEval - Cultural Reasoning - PH-Eval (Zero-Shot) Leaderboard

SeaEval - Cultural Reasoning - PH-Eval (Zero-Shot) — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 25 models tracked.

#	Model	Score
1	Qwen 2.5 72B Instruct	72
2	Qwen 2.5 32B Instruct	70
3	Llama 3.1 70B Instruct	68
4	Llama 3 70B Instruct	63
5	Qwen 2 72B Instruct	62
6	Qwen 2.5 14B Instruct	60
7	Gemma 2 9B (IT)	58
8	Llama 3 8B Instruct	58
9	Llama 3.1 8B Instruct	57
10	Llama 3 8B Cpt Sea Lionv2.1 Instruct	56
11	Qwen 2.5 7B Instruct	55
12	Sailor2-8B-Chat	53
13	Qwen 2 7B Instruct	52
14	SeaLLMs-v3-7B-Chat	47
15	Gemma 2 2B (IT)	40