List your company
ResearchFormerly WizWand

Instruction Following on Vicuna-bench

8.24ScoreLlama-2-7B-star_instruct
1.41763.18884.966.7312Nov 21, 2024Jan 31, 2025Apr 13, 2025Jun 24, 2025Sep 4, 2025Nov 15, 2025Jan 26, 2026
Updated 8mo ago

Evaluation results

MethodLinks
Llama-2-7B-star_instructModel Scale=7B2024.118.24
Llama-2-7B-evol instructModel Scale=7B2024.117.27
JREGModel Size=3.4B, Alpha (α)=1.0, SFT=true2026.016.36
Llama-2-7B-alpacaModel Scale=7B2024.116.33
Pythia-1B-star_instructModel Scale=1B2024.115.93
BaselineModel Size=3.4B, SFT=true2026.015.89
Pythia-1B-RandomModel Scale=1B2024.115.13
Pythia-1B-evol_instructModel Scale=1B2024.115.07
Pythia-1B-IFDModel Scale=1B2024.114.6
Pythia-1B-alpacaModel Scale=1B2024.114.14
Sheared-LLaMA-1.3BModel Scale=1B2024.112.73
OPT-1.3BModel Scale=1B2024.112.49
Pythia-1BModel Scale=1B2024.111.68