llama-index-commoncompute 0.1.0

LlamaIndex integration for Common Compute — Apple-Silicon-powered LLMs and embeddings for your LlamaIndex pipeline.

llama-index-commoncompute

LlamaIndex integration for Common Compute.

pip install llama-index-commoncompute

from llama_index.core import Settings, VectorStoreIndex, SimpleDirectoryReader
from llama_index_commoncompute import CommonComputeLLM, CommonComputeEmbedding

Settings.llm = CommonComputeLLM(model="qwen-2.5-7b")
Settings.embed_model = CommonComputeEmbedding(model="bge-small")

docs = SimpleDirectoryReader("./docs").load_data()
index = VectorStoreIndex.from_documents(docs)
print(index.as_query_engine().query("What workloads does Common Compute support?"))

Set CC_API_KEY in your environment (or pass api_key=... to either class).