List your company

General language understanding and reasoning on Open LLM Leaderboard v1 (test)

44.97MMLU-ProSimPO
36.02638.34840.6742.992May 16, 2025

Evaluation results

MethodLinks
SimPOBackbone=Qwen2.5-7B-Instruct, Mode=short2025.0544.9773.554.9438.631.4648.69
DPOBackbone=Qwen2.5-7B-Instruct, Mode=short2025.0544.9175.0654.9939.1231.2149.06
DPOBackbone=Qwen2.5-7B-Instruct, Mode=long2025.0544.9175.355.0637.9931.849.01
SoLo-ORPOBackbone=Qwen2.5-7B-Instruct, Mode=SoLo2025.0544.8375.1855.2337.1631.4648.77
SFTBackbone=Qwen2.5-7B-Instruct, Mode=short2025.0544.8172.1855.1536.7132.1248.19
LongPO(reimp)Backbone=Qwen2.5-7B-Instruct2025.0544.873.8655.0734.8131.9148.08
ORPOBackbone=Qwen2.5-7B-Instruct, Mode=short2025.0544.7874.755.2638.2130.9548.78
SoLo-SimPOBackbone=Qwen2.5-7B-Instruct, Mode=SoLo2025.0544.7875.954.8937.7631.5448.97
SFTBackbone=Qwen2.5-7B-Instruct, Mode=long2025.0544.7471.4655.3536.5531.4547.9
SimPOBackbone=Qwen2.5-7B-Instruct, Mode=long2025.0544.7473.555.2937.6132.0548.64
LongPO(pub)Backbone=Qwen2.5-7B-Instruct2025.0544.6976.4953.9432.3231.8747.86
SoLo-DPOBackbone=Qwen2.5-7B-Instruct, Mode=SoLo2025.0544.6673.9854.7835.5731.6348.12
ORPOBackbone=Qwen2.5-7B-Instruct, Mode=long2025.0544.6473.6155.3737.1632.1248.58
InstructBackbone=Qwen2.5-7B-Instruct2025.0544.6374.2255.2536.8631.8848.56
InstructBackbone=Llama3.1-8B-Instruct2025.0537.7962.2350.9815.2531.7139.59
SoLo-ORPOBackbone=Llama3.1-8B-Instruct, Mode=SoLo2025.0537.663.4350.1815.6331.3839.64
ORPOBackbone=Llama3.1-8B-Instruct, Mode=long2025.0537.4360.4350.8615.8631.6339.24
ORPOBackbone=Llama3.1-8B-Instruct, Mode=short2025.0537.2661.1550.9514.8832.1339.27
SFTBackbone=Llama3.1-8B-Instruct, Mode=long2025.0537.0460.6749.6416.1632.3839.17
SFTBackbone=Llama3.1-8B-Instruct, Mode=short2025.0536.3760.3150.2317.931.7939.32