List your company
ResearchFormerly WizWand

Accuracy on PiQA

76.56AccuracyCoA-LoRA
52.005658.380364.75571.1297Jul 21, 2025Aug 31, 2025Oct 12, 2025Nov 22, 2025Jan 3, 2026Feb 13, 2026Mar 27, 2026
Updated 6mo ago

Evaluation results

MethodLinks
CoA-LoRAModel=Llama-2-7B, Average Bit Budget=3-bit, Zero-shot=true2025.0976.56
Base ModelBackbone=LLaMA-DCLM, Reweighting Space=Data sources2025.0775.84
Base ModelBackbone=LLaMA-Nemo, Reweighting Space=Data sources2025.0775.57
LQ-LoRAModel=Llama-2-7B, Average Bit Budget=3-bit, Zero-shot=true2025.0975
DataAgent_RLBackbone=LLaMA-DCLM, Reweighting Space=Data sources2025.0774.85
DataAgent_SFTBackbone=LLaMA-FWE, Reweighting Space=Data sources2025.0774.6
DataAgent_RLBackbone=LLaMA-Nemo, Reweighting Space=Data sources2025.0774.49
Shared-LoRAModel=Llama-2-7B, Average Bit Budget=3-bit, Zero-shot=true2025.0974.22
Base ModelBackbone=LLaMA-FWE, Reweighting Space=Data sources2025.0774.21
DataAgent_SFTBackbone=LLaMA-Nemo, Reweighting Space=Data sources2025.0774.17
RegMixBackbone=LLaMA-DCLM, Reweighting Space=Data sources2025.0773.88
QLoRAModel=Llama-2-7B, Average Bit Budget=3-bit, Zero-shot=true2025.0973.44
Apertus-8BModel Category=Fully open models, Instruction-tuned=true2026.0373.4
DataAgent_SFTBackbone=LLaMA-DCLM, Reweighting Space=Data sources2025.0773.28
DataAgent_RLBackbone=LLaMA-FWE, Reweighting Space=Data sources2025.0773.12
RegMixBackbone=LLaMA-Nemo, Reweighting Space=Data sources2025.0772.85
AMALIA-9B-DPOModel Category=Fully open models, Training=DPO, Instruction-tuned=true2026.0372.5
DBLBackbone=LLaMA-DCLM, Reweighting Space=Data sources2025.0772.5
Shared-LoRAModel=Qwen2.5-3B, Average Bit Budget=3-bit, Zero-shot=true2025.0971.88
RegMixBackbone=LLaMA-FWE, Reweighting Space=Data sources2025.0771.72
EuroLLM-9BModel Category=Fully open models, Instruction-tuned=true2026.0371.7
AMALIA-9B-SFTModel Category=Fully open models, Training=SFT, Instruction-tuned=true2026.0371.7
DBLBackbone=LLaMA-FWE, Reweighting Space=Data sources2025.0771.18
Base ModelBackbone=Pythia-1.4B, Reweighting Space=Data sources2025.0770.89
DBLBackbone=LLaMA-Nemo, Reweighting Space=Data sources2025.0770.79
Ministral-8BModel Category=Open weight models, Instruction-tuned=true2026.0370.7
Salamandra-7BModel Category=Fully open models, Instruction-tuned=true2026.0370.4
QLoRAModel=Qwen2.5-3B, Average Bit Budget=3-bit, Zero-shot=true2025.0970.34
CoA-LoRAModel=Qwen2.5-3B, Average Bit Budget=3-bit, Zero-shot=true2025.0970.31
Gemma 2-9BModel Category=Open weight models, Instruction-tuned=true2026.0370.3
Gervasio-8BModel Category=Open weight models, Instruction-tuned=true2026.0369.1
Gemma 3-12BModel Category=Open weight models, Instruction-tuned=true2026.0369
LQ-LoRAModel=Qwen2.5-3B, Average Bit Budget=3-bit, Zero-shot=true2025.0968.75
Llama 3.1-8BModel Category=Open weight models, Instruction-tuned=true2026.0368.5
Mistral-7BModel Category=Open weight models, Instruction-tuned=true2026.0366.9
Qwen 2.5-7BModel Category=Open weight models, Instruction-tuned=true2026.0366.6
RegMixBackbone=Pythia-1.4B, Reweighting Space=Data sources2025.0765.7
DataAgent_RLBackbone=Pythia-1.4B, Reweighting Space=Data sources2025.0765.54
Qwen 3-8BModel Category=Open weight models, Instruction-tuned=true2026.0364.7
DataAgent_SFTBackbone=Pythia-1.4B, Reweighting Space=Data sources2025.0761.94
Naive TrainingBackbone=LLaMA-FWE, Reweighting Space=Data sources2025.0761.1
OLMo 2-7BModel Category=Fully open models, Instruction-tuned=true2026.0361
Naive TrainingBackbone=LLaMA-DCLM, Reweighting Space=Data sources2025.0760.28
Naive TrainingBackbone=LLaMA-Nemo, Reweighting Space=Data sources2025.0759.74
Naive TrainingBackbone=Pythia-1.4B, Reweighting Space=Data sources2025.0752.95