List your company
ResearchFormerly WizWand

Safety Evaluation on SST (ASR)

20.4ASROLMo 2-7B
-0.4510.415.8Mar 27, 2026
Updated 7mo ago

Evaluation results

MethodLinks
OLMo 2-7BModel category=Fully open models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.0320.4
Mistral-7BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.0319.4
Salamandra-7BModel category=Fully open models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.0318.4
Gervasio-8BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.039.4
Ministral-8BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.038.4
Llama 3.1-8BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.037.4
Apertus-8BModel category=Fully open models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.034.4
Qwen 3-8BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.034.4
EuroLLM-9BModel category=Fully open models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.033.4
AMALIA-9B-SFTModel category=Fully open models, Instruction-tuned=true, Post-training method=SFT, Judge model=Qwen3Guard-8B2026.033.4
Qwen 2.5-7BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.033.4
Gemma 3-12BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.032.4
AMALIA-9B-DPOModel category=Fully open models, Instruction-tuned=true, Post-training method=DPO, Judge model=Qwen3Guard-8B2026.031.4
Gemma 2-9BModel category=Open weight models, Instruction-tuned=true, Judge model=Qwen3Guard-8B2026.030.4