Ask a question to get started
Enter to send•Shift+Enter new line
BigdlLLM()
IpexLLM
Model name or model path to use.
Keyword arguments passed to the model.
IpexLLM model.
Huggingface tokenizer model.
Whether to stream the results, token by token.
Construct object from model_id
Construct low_bit object from model_id
Args:
model_id: Path for the bigdl-llm transformers low-bit model folder. tokenizer_id: Path for the huggingface repo id or local model folder which contains the tokenizer. model_kwargs: Keyword arguments to pass to the model and tokenizer. kwargs: Extra arguments to pass to the model and tokenizer.
Wrapper around the BigdlLLM model
Example:
.. code-block:: python
from langchain_community.llms import BigdlLLM llm = BigdlLLM.from_model_id(model_id="THUDM/chatglm-6b")
