Deploy BAAR2-9M: Authentic tokenizer, SafeTensors and pure INT8/INT4 .baar2 models

Delete vlite_manifest.json with huggingface_hub

Delete model_shared_int4.vlite with huggingface_hub

Delete model_shared_int3.vlite with huggingface_hub

Delete model_shared_int2.vlite with huggingface_hub

Delete model_shared_int1.vlite with huggingface_hub

Delete model_shared.vlite with huggingface_hub

Delete model_micro_int4.vlite with huggingface_hub

Delete model_micro_int3.vlite with huggingface_hub

Delete model_micro_int2.vlite with huggingface_hub

Delete model_micro_int1.vlite with huggingface_hub

Delete model_full_int4.vlite with huggingface_hub

Delete model_full_int3.vlite with huggingface_hub

Delete model_full_int2.vlite with huggingface_hub

Delete model_full_int1.vlite with huggingface_hub

Delete model.vlite with huggingface_hub

Deploy 100% pure INT8 9M model .baar2 (9.21 MB)

Deploy 100% pure INT8 9M model .baar2 (9.21 MB)

Update README.md

Update README.md

Create README.md

Deploy 100% pure INT4 9M model .baar2 (4.65 MB)

Deploy 100% pure INT4 full performance .baar2 (4.65 MB)

Upload folder using huggingface_hub

Upload checkpoint_step_50000.pt with huggingface_hub

Upload checkpoint_step_50000.pt with huggingface_hub

Upload checkpoint_step_49000.pt with huggingface_hub

initial commit