SLIM-EXTRACT-PHI-3-GGUF

slim-extract-phi-3-gguf is a function-calling extraction, finetuned on top of phi-3-mini, and then then 4_K_M quantized with GGUF, providing a small, fast inference implementation, optimized for multi-model concurrent deployment.

This model has been fine-tuned to implement a general-purpose extraction function that takes a custom key as input parameter, and generates a python dictionary consisting of that custom key with the value consisting of a list of the values associated with that key in the text.

The intent of SLIMs is to forge a middle-ground between traditional encoder-based classifiers and open-ended API-based LLMs.

slim-extract is part of the SLIM ("Structured Language Instruction Model") series, providing a set of small, specialized decoder-based LLMs, fine-tuned for function-calling.

To pull the model via API:

from huggingface_hub import snapshot_download           
snapshot_download("llmware/slim-extract-phi-3-gguf", 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-extract-phi-3-gguf")
response = model.function_call(text_sample)  

# this one line will download the model and run a series of tests
ModelCatalog().tool_test_run("slim-extract-phi-3-gguf", verbose=True)  

Note: please review config.json in the repository for prompt wrapping information, details on the model, and full test set.

Model Card Contact

Darren Oberst & llmware team

Any questions? Join us on Discord