List your company

Machine Translation on WMT23 (test)

83.2COMET ScoreX-ALMA-13B (Only SFT, MoE)
70.7273.9677.280.44Feb 16, 2025

Evaluation results

MethodLinks
X-ALMA-13B (Only SFT, MoE)# Tokens (Pre-training)=110B, # Params (Base/Adapter)=13B/16B, # Params (Total)=29B2025.0283.2————————————
X-ALMA-13B-DAT (MoE)# Tokens (Pre-training)=20B, # Params (Base/Adapter)=13B/4B, # Params (Total)=17B2025.0282.8————————————
Aya-23-35B# Tokens (Pre-training)=null, # Params (Base/Adapter)=35B/-, # Params (Total)=35B2025.0282.3————————————
X-ALMA-13B-DATM (MoE)# Tokens (Pre-training)=20B, # Params (Base/Adapter)=13B/1B, # Params (Total)=14B2025.0282.1————————————
Aya-23-8B# Tokens (Pre-training)=null, # Params (Base/Adapter)=8B/-, # Params (Total)=8B2025.0281.5————————————
LLaMAX3-Alpaca-8B# Tokens (Pre-training)=66B, # Params (Base/Adapter)=8B/-, # Params (Total)=8B2025.0281————————————
Aya-101# Tokens (Pre-training)=null, # Params (Base/Adapter)=13B/-, # Params (Total)=13B2025.0279.7————————————
NLLB-3.3B# Tokens (Pre-training)=null, # Params (Base/Adapter)=3B/-, # Params (Total)=3B2025.0271.2————————————
Aya 23 35BParameters=35B2025.02—23.579.732.782.331.782.221.381.639.185.729.782.3
Aya 23 8BParameters=8B2025.02—22.678.832.382.130.981.619.880.239.2852981.5
Aya-101Parameters=13B2025.02—13.873.734.981.628.481.413.977.334.984.525.279.7
LLaMAX3-Alpaca-8BBackbone=LLaMA-3, Parameters=8B2025.02—22.379.325.679.429.481.317.680.137.884.926.581
NLLB-3.3BParameters=3.3B2025.02—11.467.820.166.624.476.76.865.833.17919.271.2
X-ALMA-13B (Only SFT, MoE)Architecture=MoE, Parameters=13B, Training Mode=Only SFT2025.02—23.880.342.585.332.882.420.481.642.586.432.483.2
X-ALMA-13B-DAT (MoE)Architecture=MoE, Parameters=13B, Training Mode=DAT2025.02—21.179.738.484.632.382.419.881.239.585.930.282.8
X-ALMA-13B-DATM (MoE)Architecture=MoE, Parameters=13B, Training Mode=DATM2025.02—22.279.237.584.229821980.43484.728.382.1