llama-test-model / serving.properties
silviachen46's picture
Upload 17 files
38020f8
raw
history blame contribute delete
205 Bytes
engine=DeepSpeed
option.entryPoint=inference.py
option.tensor_parallel_degree=1
option.max_tokens=4096
option.task=text-generation
option.dtype=bf16
option.model_type=base
option.model_loading_timeout=3600