List your company

Instruction Following on Alpaca Finance

2.78Average LengthOpt-Hydra
1.07441.51721.962.4028Mar 13, 2026

Evaluation results

MethodLinks
Opt-HydraTarget model=meta-llama / Llama-2-7B-Chat2026.032.781.68
Opt-HydraTarget model=lmsys / Vicuna-7B-v1.32026.032.71.55
OSD-HydraTarget model=meta-llama / Llama-2-7B-Chat2026.032.251.36
OSD-HydraTarget model=lmsys / Vicuna-7B-v1.32026.032.191.25
Ens-EAGLE-3Target model=meta-llama / Llama-2-7B-Chat2026.032.151.26
OSD-EAGLE-3Target model=meta-llama / Llama-2-7B-Chat2026.032.131.23
Ens-EAGLE-3Target model=lmsys / Vicuna-7B-v1.32026.032.071.27
OSD-EAGLE-3Target model=lmsys / Vicuna-7B-v1.32026.0321.16
HydraTarget model=lmsys / Vicuna-7B-v1.32026.031.781
EAGLE-3Target model=meta-llama / Llama-2-7B-Chat2026.031.71
HydraTarget model=meta-llama / Llama-2-7B-Chat2026.031.641
EAGLE-3Target model=lmsys / Vicuna-7B-v1.32026.031.621
Ens-EAGLETarget model=lmsys / Vicuna-7B-v1.32026.031.611.14
Ens-EAGLETarget model=meta-llama / Llama-2-7B-Chat2026.031.561.41
OSD-EAGLETarget model=lmsys / Vicuna-7B-v1.32026.031.511.09
OSD-EAGLETarget model=meta-llama / Llama-2-7B-Chat2026.031.431.2
EAGLETarget model=lmsys / Vicuna-7B-v1.32026.031.391
OSDTarget model=meta-llama / Llama-2-7B-Chat2026.031.311.12
OSDTarget model=lmsys / Vicuna-7B-v1.32026.031.31.08
Vanilla SDTarget model=lmsys / Vicuna-7B-v1.32026.031.21
Vanilla SDTarget model=meta-llama / Llama-2-7B-Chat2026.031.21
EAGLETarget model=meta-llama / Llama-2-7B-Chat2026.031.141