List your company
ResearchFormerly WizWand

Leaderboard Evaluation on Open LLM Leaderboard 2 (Overall Score)

51.28Overall ScoreUltraMix-190k
25.0231.837538.65545.4725Nov 14, 2025
Updated 8mo ago

Evaluation results

MethodLinks
UltraMix-190kBase Model=Qwen-2.5-7B-TuluSFT, Training Mixture/Protocol=UM-190k2025.1151.28
UltraMix-187kBase Model=Qwen-2.5-7B-TuluSFT, Training Mixture/Protocol=UM-187k2025.1150.83
UltraMix-170kBase Model=Qwen-2.5-7B-TuluSFT, Training Mixture/Protocol=UM-170k2025.1149.74
TuluDPOBase Model=Qwen-2.5-7B-TuluSFT, Training Mixture/Protocol=TuluDPO2025.1148.53
SFTBase Model=Qwen-2.5-7B-TuluSFT, Training Mixture/Protocol=SFT2025.1145.77
UltraMix-190kBase Model=Llama-3.1-8B-TuluSFT, Training Mixture/Protocol=UM-190k2025.1142.75
UltraMixTraining Method=DPO, Dataset=UM-190k2025.1142.06
UltraMix-187kBase Model=Llama-3.1-8B-TuluSFT, Training Mixture/Protocol=UM-187k2025.1141.92
UltraMixTraining Method=DPO, Dataset=UM-187k2025.1141.65
UltraMix-170kBase Model=Llama-3.1-8B-TuluSFT, Training Mixture/Protocol=UM-170k2025.1141.02
TuluDPOTraining Method=DPO, Dataset=TuluDPO2025.1140.68
TuluDPOBase Model=Llama-3.1-8B-TuluSFT, Training Mixture/Protocol=TuluDPO2025.1140.66
UltraMixTraining Method=DPO, Dataset=UM-170k2025.1140.28
ORPOTraining Method=ORPO, Dataset=ORPO2025.1139.91
Instella-3B-SFT (UM-190k)DPO Training Dataset=UM-190k2025.1138.42
UM-190kBackbone=Apertus-8B-SFT2025.1138.36
Instella-3B-SFT (UM-187k)DPO Training Dataset=UM-187k2025.1138.01
UM-187kBackbone=Apertus-8B-SFT2025.1137.93
SmolLM-3-3B-SFTTraining Method=SFT2025.1137.72
Instella-3B-SFT (UM-170k)DPO Training Dataset=UM-170k2025.1137.32
SFTBase Model=Llama-3.1-8B-TuluSFT, Training Mixture/Protocol=SFT2025.1137.15
Instella-3B-SFT (TuluDPO)DPO Training Dataset=TuluDPO2025.1137.14
UM-170kBackbone=Apertus-8B-SFT2025.1136.94
TuluDPOBackbone=Apertus-8B-SFT2025.1136.76
HelpSteerTraining Method=DPO, Dataset=HelpSteer2025.1136.5
UltraFBBackbone=Apertus-8B-SFT2025.1136.36
Instella-3B-SFT (ORPO)DPO Training Dataset=ORPO2025.1136.25
ORPOBackbone=Apertus-8B-SFT2025.1135.95
UltraFBTraining Method=DPO, Dataset=UltraFB2025.1135.94
HelpSteerBackbone=Apertus-8B-SFT2025.1135.43
Instella-3B-SFT (HelpSteer)DPO Training Dataset=HelpSteer2025.1135.11
Instella-3B-SFT (SFT)DPO Training Dataset=None (SFT Baseline)2025.1134.89
Instella-3B-SFT (UltraFB)DPO Training Dataset=UltraFB2025.1134.88
CodePrefTraining Method=DPO, Dataset=CodePref2025.1134.7
Instella-3B-SFT (CodePref)DPO Training Dataset=CodePref2025.1134.38
SFTBackbone=Apertus-8B-SFT2025.1134.21
CodePrefBackbone=Apertus-8B-SFT2025.1134.02
UM-190kTraining Dataset=UM-190k2025.1131.07
UM-187kTraining Dataset=UM-187k2025.1130.91
UM-190k2025.1130.67
UM-170kTraining Dataset=UM-170k2025.1130.18
UM-187k2025.1130.15
TuluDPOTraining Dataset=TuluDPO2025.1130.02
TuluDPO2025.1129.3
ORPOTraining Dataset=ORPO2025.1129.13
ORPO2025.1129.13
UM-170k2025.1128.64
UltraFBTraining Dataset=UltraFB2025.1128.52
SFT2025.1128.39
UltraFB2025.1128.2
HelpSteerTraining Dataset=HelpSteer2025.1127.85
CodePrefTraining Dataset=CodePref2025.1127.65
HelpSteer2025.1127.56
SFTTraining Dataset=SFT2025.1127.31
CodePref2025.1126.03