List your company

General Knowledge on MMLU-Pro (pass@1)

66.46pass@1Qwen3-8B
28.572838.408948.24558.0811Jan 8, 2026Jan 26, 2026Feb 13, 2026Mar 3, 2026Mar 21, 2026Apr 8, 2026Apr 26, 2026

Evaluation results

MethodLinks
Qwen3-8BBackbone=Qwen3-8B2026.0166.46
RelayLLM (Difficulty-Aware)Student Model Backbone=Qwen3-1.7B2026.0159.03
RelayLLM (Simple)Student Model Backbone=Qwen3-1.7B2026.0158.76
SRFT [13]Base Model=Qwen2.5-Math-7B, Source=reported2026.0455.9
SFT→RLBase Model=Qwen2.5-Math-7B, Source=verl2026.0455.1
ReLIFT [12]Base Model=Qwen2.5-Math-7B, Source=reported2026.0453.9
CITERStudent Model Backbone=Qwen3-1.7B2026.0153.38
LUFFY [11]Base Model=Qwen2.5-Math-7B, Source=reported2026.0453
HPT [15]Base Model=Qwen2.5-Math-7B, Source=reported2026.0452.5
Prefix-RFT [14]Base Model=Qwen2.5-Math-7B, Source=reported2026.0452.1
ReLIFT [12]Base Model=Qwen2.5-Math-7B, Source=reproduction attempt2026.0451.3
LUFFY [11]Base Model=Qwen2.5-Math-7B, Source=reproduction attempt2026.0450.9
SFTBase Model=Qwen2.5-Math-7B, Source=verl2026.0450.2
GRPOStudent Model Backbone=Qwen3-1.7B2026.0149.76
Base ModelStudent Model Backbone=Qwen3-1.7B2026.0146.9
RelayLLM (Difficulty-Aware)Student Model Backbone=Qwen3-0.6B2026.0135.87
RelayLLM (Simple)Student Model Backbone=Qwen3-0.6B2026.0135.61
CITERStudent Model Backbone=Qwen3-0.6B2026.0133.12
GRPOStudent Model Backbone=Qwen3-0.6B2026.0132.15
Base ModelStudent Model Backbone=Qwen3-0.6B2026.0130.03