Our new X account is live! Follow @wizwand_team for updates

Multi-level multi-discipline evaluation on C-Eval

81.4Accuracy

Yi

Updated 4d ago

Evaluation Results

Method	Links
Yi 2024.03		81.4
Qwen 2024.03		72.1
Yi 2024.03		72
GPT-4 2024.03		69.9
Qwen2.5-1.5B 2026.02		68.2
Qwen3-1.7B 2026.02		65.2
Skywork 2024.03		60.6
Baichuan-2 2024.03		59
InternLM 2024.03		58.8
Falcon 2024.03		57.8
GPT-3.5 2024.03		52.5
Qwen2.5-0.5B 2026.02		51
Qwen3-0.6B 2026.02		50.4
Llama2 2024.03		50.1
SPES-9B 2026.02		44.7
SmolLM2-1.7B 2026.02		32.5
OLMoE-1B-7B 2026.02		31.1
Llama3.2-1B 2026.02		30.9
MobiLlama-1.3B 2026.02		26.2
SPES-7B 2026.02		26.2
TinyLlama-1.1B 2026.02		26
SPES-2B 2026.02		25
MoE++ 7B 2026.02		23.6
OPT-1.3B 2026.02		23
Pythia-1.4B 2026.02		23
OPT-2.7B 2026.02		23
Pythia-2.8B 2026.02		22.9
MobiLlama-0.8B 2026.02		22.7