Skip to content

armgit / LEANN

Public Imported from GitHub
0

[MLsys2026 Best Paper]: https://arxiv.org/abs/2506.08276. RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.

rm useless

yichuan52003091032095a653993a9a
diff --git a/examples/main_cli_example.py b/examples/main_cli_example.py
index 262fa94..e89e1ee 100644
--- a/examples/main_cli_example.py
+++ b/examples/main_cli_example.py
@@ -23,7 +23,7 @@ file_extractor: dict[str, BaseReader] = {
".xlsx": reader,
}
node_parser = DoclingNodeParser(
- chunker=HybridChunker(tokenizer="Qwen/Qwen3-Embedding-4B", max_tokens=64)
+ chunker=HybridChunker(tokenizer="Qwen/Qwen3-Embedding-4B", max_tokens=128)
)
print("Loading documents...")
documents = SimpleDirectoryReader(
diff --git a/packages/leann-core/src/leann/api.py b/packages/leann-core/src/leann/api.py
index ed6af41..48478bf 100644
--- a/packages/leann-core/src/leann/api.py
+++ b/packages/leann-core/src/leann/api.py
@@ -32,6 +32,7 @@ def _compute_embeddings(chunks: List[str], model_name: str) -> np.ndarray:
else:
from sentence_transformers import SentenceTransformer
model = SentenceTransformer(model_name)
+ model = model.half()
print(f"INFO: Computing embeddings for {len(chunks)} chunks using SentenceTransformer model '{model_name}'...")
embeddings = model.encode(chunks, show_progress_bar=True)