[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.
Python aifaissgpt-osslangchainllama-indexllmlocalstorageoffline-firstollamaprivacypythonragretrieval-augmented-generationvector-databasevector-searchvectors
fix: logs
Andy Leec2f35c8e73dd
diff --git a/packages/leann-backend-hnsw/leann_backend_hnsw/hnsw_embedding_server.py b/packages/leann-backend-hnsw/leann_backend_hnsw/hnsw_embedding_server.pyindex 94b6529..07e74c4 100644--- a/packages/leann-backend-hnsw/leann_backend_hnsw/hnsw_embedding_server.py+++ b/packages/leann-backend-hnsw/leann_backend_hnsw/hnsw_embedding_server.py@@ -17,12 +17,20 @@ import logging# Set up logging based on environment variableLOG_LEVEL = os.getenv("LEANN_LOG_LEVEL", "WARNING").upper()-logging.basicConfig(- level=getattr(logging, LOG_LEVEL, logging.INFO),- format="%(asctime)s - %(levelname)s - %(message)s",-)logger = logging.getLogger(__name__)+# Force set logger level (don't rely on basicConfig in subprocess)+log_level = getattr(logging, LOG_LEVEL, logging.WARNING)+logger.setLevel(log_level)++# Ensure we have a handler if none exists+if not logger.handlers:+ handler = logging.StreamHandler()+ formatter = logging.Formatter("%(asctime)s - %(levelname)s - %(message)s")+ handler.setFormatter(formatter)+ logger.addHandler(handler)+ logger.propagate = False+def create_hnsw_embedding_server(passages_file: Optional[str] = None,diff --git a/packages/leann-core/src/leann/embedding_compute.py b/packages/leann-core/src/leann/embedding_compute.pyindex 38dde43..d0df463 100644--- a/packages/leann-core/src/leann/embedding_compute.py+++ b/packages/leann-core/src/leann/embedding_compute.py@@ -55,8 +55,8 @@ def compute_embeddings_sentence_transformers("""Compute embeddings using SentenceTransformer with model caching"""- print(- f"INFO: Computing embeddings for {len(texts)} texts using SentenceTransformer, model: '{model_name}'"+ logger.info(+ f"Computing embeddings for {len(texts)} texts using SentenceTransformer, model: '{model_name}'")# Auto-detect device@@ -73,13 +73,13 @@ def compute_embeddings_sentence_transformers(# Check if model is already cachedif cache_key in _model_cache:- print(f"INFO: Using cached model: {model_name}")+ logger.info(f"Using cached model: {model_name}")model = _model_cache[cache_key]else:- print(f"INFO: Loading and caching SentenceTransformer model: {model_name}")+ logger.info(f"Loading and caching SentenceTransformer model: {model_name}")from sentence_transformers import SentenceTransformer- print(f"INFO: Using device: {device}")+ logger.info(f"Using device: {device}")# Prepare model and tokenizer optimization parametersmodel_kwargs = {@@ -104,9 +104,9 @@ def compute_embeddings_sentence_transformers(tokenizer_kwargs=tokenizer_kwargs,local_files_only=True,)- print("✅ Model loaded successfully! (local + optimized)")+ logger.info("Model loaded successfully! (local + optimized)")except Exception as e:- print(f"Local loading failed ({e}), trying network download...")+ logger.warning(f"Local loading failed ({e}), trying network download...")# Fallback to network loadingmodel_kwargs["local_files_only"] = Falsetokenizer_kwargs["local_files_only"] = False@@ -118,23 +118,23 @@ def compute_embeddings_sentence_transformers(tokenizer_kwargs=tokenizer_kwargs,local_files_only=False,)- print("✅ Model loaded successfully! (network + optimized)")+ logger.info("Model loaded successfully! (network + optimized)")# Apply additional optimizations (if supported)if use_fp16 and device in ["cuda", "mps"]:try:model = model.half()model = torch.compile(model)- print(f"✅ Using FP16 precision and compile optimization: {model_name}")+ logger.info(f"Using FP16 precision and compile optimization: {model_name}")except Exception as e:- print(f"FP16 or compile optimization failed: {e}")+ logger.warning(f"FP16 or compile optimization failed: {e}")# Cache the model_model_cache[cache_key] = model- print(f"✅ Model cached: {cache_key}")+ logger.info(f"Model cached: {cache_key}")# Compute embeddings- print("INFO: Starting embedding computation...")+ logger.info("Starting embedding computation...")embeddings = model.encode(texts,@@ -178,10 +178,10 @@ def compute_embeddings_openai(texts: List[str], model_name: str) -> np.ndarray:else:client = openai.OpenAI(api_key=api_key)_model_cache[cache_key] = client- print("✅ OpenAI client cached")+ logger.info("OpenAI client cached")- print(- f"INFO: Computing embeddings for {len(texts)} texts using OpenAI API, model: '{model_name}'"+ logger.info(+ f"Computing embeddings for {len(texts)} texts using OpenAI API, model: '{model_name}'")# OpenAI has limits on batch size and input length@@ -208,7 +208,7 @@ def compute_embeddings_openai(texts: List[str], model_name: str) -> np.ndarray:batch_embeddings = [embedding.embedding for embedding in response.data]all_embeddings.extend(batch_embeddings)except Exception as e:- print(f"ERROR: Batch {i} failed: {e}")+ logger.error(f"Batch {i} failed: {e}")raiseembeddings = np.array(all_embeddings, dtype=np.float32)@@ -231,20 +231,20 @@ def compute_embeddings_mlx("MLX or related libraries not available. Install with: uv pip install mlx mlx-lm") from e- print(- f"INFO: Computing embeddings for {len(chunks)} chunks using MLX model '{model_name}' with batch_size={batch_size}..."+ logger.info(+ f"Computing embeddings for {len(chunks)} chunks using MLX model '{model_name}' with batch_size={batch_size}...")# Cache MLX model and tokenizercache_key = f"mlx_{model_name}"if cache_key in _model_cache:- print(f"INFO: Using cached MLX model: {model_name}")+ logger.info(f"Using cached MLX model: {model_name}")model, tokenizer = _model_cache[cache_key]else:- print(f"INFO: Loading and caching MLX model: {model_name}")+ logger.info(f"Loading and caching MLX model: {model_name}")model, tokenizer = load(model_name)_model_cache[cache_key] = (model, tokenizer)- print(f"✅ MLX model cached: {cache_key}")+ logger.info(f"MLX model cached: {cache_key}")# Process chunks in batches with progress barall_embeddings = []