TinyLlama/TinyLlama-1.1B-Chat-v1.0
Text Generation • 1B • Updated • 3.24M • • 1.58k
Models for LaaS
Collection of models that we are interested in running. Categorized by: (1) Text generation for inference, (2) Smaller models that we want to FT
-
-
tiiuae/falcon-7b-instruct
Text Generation • 7B • Updated • 52.5k • 1.03k -
mistralai/Mistral-7B-Instruct-v0.2
Text Generation • 7B • Updated • 2.21M • • 3.13k -
meta-llama/Llama-2-7b
Text Generation • Updated • 312 • 4.48k -
microsoft/phi-2
Text Generation • 3B • Updated • 612k • 3.45k -
google-t5/t5-small
Translation • 60.5M • Updated • 2.34M • • 543
distilbert/distilgpt2
Text Generation • 88.2M • Updated • 2.81M • 625Note Text generation models.
-
google-bert/bert-base-uncased
Fill-Mask • 0.1B • Updated • 59.2M • • 2.64k
prajjwal1/bert-tiny
Updated • 2.48M • 145Note BERT models generally for fine-tuning. On inference, they are the base encoder models and only do MLM
EfficientNetV2: Smaller Models and Faster Training
Paper • 2104.00298 • Published • 1Note Replace with the relevant models later. Ideally want: EfficientNet (v1,v2) MobileNet YOLO (v_x) Resnet variants FPN Diffusion: SAM (?) or SEEM