List your company
ResearchFormerly WizWand

Language Modeling on General Capability

0.685Capability ScoreIT+RL
0.298120.398560.4990.59944Dec 29, 2025
Updated 8mo ago

Evaluation results

MethodLinks
IT+RLModel=Llama-3.1-8B, # Params=100%2025.120.685
OriginalModel=Llama-3.1-8B, # Params=-2025.120.674
DoRAModel=Llama-3.1-8B, # Params=0.22%2025.120.67
SAILSModel=Llama-3.1-8B, # Params=0.21%2025.120.67
LoRAModel=Llama-3.1-8B, # Params=0.21%2025.120.652
IT+RLModel=Gemma-2-9B, # Params=100%2025.120.57
PromptModel=Llama-3.1-8B, # Params=-2025.120.545
IT+RLModel=Gemma-2-2B, # Params=100%2025.120.495
OriginalModel=Gemma-2-2B, # Params=-2025.120.431
OriginalModel=Gemma-2-9B, # Params=-2025.120.411
SAILSModel=Gemma-2-9B, # Params=0.19%2025.120.404
PromptModel=Gemma-2-2B, # Params=-2025.120.392
PromptModel=Gemma-2-9B, # Params=-2025.120.39
SAILSModel=Gemma-2-2B, # Params=0.24%2025.120.366
DoRAModel=Gemma-2-2B, # Params=0.25%2025.120.365
LoRAModel=Gemma-2-2B, # Params=0.24%2025.120.362
LoRAModel=Gemma-2-9B, # Params=0.19%2025.120.35
DoRAModel=Gemma-2-9B, # Params=0.20%2025.120.34
FFTModel=Llama-3.1-8B, # Params=100%2025.120.339
FFTModel=Gemma-2-2B, # Params=100%2025.120.321
FFTModel=Gemma-2-9B, # Params=100%2025.120.313