slim-xsum-tool is a 4_K_M quantized GGUF version of slim-xsum, providing a small, fast inference implementation, optimized for multi-model concurrent deployment.
This model implements an 'extreme summarization' (e.g., 'xsum') function based on the parameter key "xsum" that generates an LLM text output in the form of a python dictionary as follows:
`{'xsum': ['Stock Market declines on worries of interest rates.']} `
The intent of SLIMs is to forge a middle-ground between traditional encoder-based classifiers and open-ended API-based LLMs through the use of function-calling and small specialized LLMs.
slim-xsum is the Pytorch version of the model, and suitable for fine-tuning for further domain adaptation.
To pull the model via API:
from huggingface_hub import snapshot_download
snapshot_download("llmware/slim-xsum-tool", local_dir="/path/on/your/machine/", local_dir_use_symlinks=False)
Load in your favorite GGUF inference engine, or try with llmware as follows:
from llmware.models import ModelCatalog
# to load the model and make a basic inference
model = ModelCatalog().load_model("slim-xsum-tool")
response = model.function_call(text_sample)
# this one line will download the model and run a series of tests
ModelCatalog().tool_test_run("slim-xsum-tool", verbose=True)
Note: please review config.json in the repository for prompt wrapping information, details on the model, and full test set.
