{ "cells": [ { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "INFO:sentence_transformers.SentenceTransformer:Use pytorch device_name: mps\n", "INFO:sentence_transformers.SentenceTransformer:Load pretrained SentenceTransformer: facebook/contriever-msmarco\n", "WARNING:sentence_transformers.SentenceTransformer:No sentence-transformers model found with name facebook/contriever-msmarco. Creating a new one with mean pooling.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "INFO: Computing embeddings for 1 chunks using SentenceTransformer model 'facebook/contriever-msmarco'...\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "Batches: 100%|██████████| 1/1 [00:00<00:00, 13.92it/s]\n", "INFO:sentence_transformers.SentenceTransformer:Use pytorch device_name: mps\n", "INFO:sentence_transformers.SentenceTransformer:Load pretrained SentenceTransformer: facebook/contriever-msmarco\n", "WARNING:sentence_transformers.SentenceTransformer:No sentence-transformers model found with name facebook/contriever-msmarco. Creating a new one with mean pooling.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "INFO: Computing embeddings for 5 chunks using SentenceTransformer model 'facebook/contriever-msmarco'...\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "Batches: 100%|██████████| 1/1 [00:00<00:00, 50.18it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ "M: 64 for level: 0\n", "INFO: Converting HNSW index to CSR-pruned format...\n", "Starting conversion: knowledge.index -> knowledge.csr.tmp\n", "[0.00s] Reading Index HNSW header...\n", "[0.00s] Header read: d=768, ntotal=5\n", "[0.00s] Reading HNSW struct vectors...\n", " Reading vector (dtype=, fmt='d')... Count=6, Bytes=48\n", "[0.00s] Read assign_probas (6)\n", " Reading vector (dtype=, fmt='i')... Count=7, Bytes=28\n", "[0.13s] Read cum_nneighbor_per_level (7)\n", " Reading vector (dtype=, fmt='i')... " ] }, { "name": "stderr", "output_type": "stream", "text": [ "\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Count=5, Bytes=20\n", "[0.23s] Read levels (5)\n", "[0.31s] Probing for compact storage flag...\n", "[0.31s] Found compact flag: False\n", "[0.31s] Compact flag is False, reading original format...\n", "[0.31s] Probing for potential extra byte before non-compact offsets...\n", "[0.31s] Found and consumed an unexpected 0x00 byte.\n", " Reading vector (dtype=, fmt='Q')... Count=6, Bytes=48\n", "[0.31s] Read offsets (6)\n", "[0.40s] Attempting to read neighbors vector...\n", " Reading vector (dtype=, fmt='i')... Count=320, Bytes=1280\n", "[0.40s] Read neighbors (320)\n", "[0.49s] Read scalar params (ep=4, max_lvl=0)\n", "[0.49s] Checking for storage data...\n", "[0.49s] Found storage fourcc: 49467849.\n", "[0.49s] Converting to CSR format...\n", "[0.49s] Conversion loop finished. \n", "[0.49s] Running validation checks...\n", " Checking total valid neighbor count...\n", " OK: Total valid neighbors = 20\n", " Checking final pointer indices...\n", " OK: Final pointers match data size.\n", "[0.49s] Deleting original neighbors and offsets arrays...\n", " CSR Stats: |data|=20, |level_ptr|=10\n", "[0.57s] Writing CSR HNSW graph data in FAISS-compatible order...\n", " Pruning embeddings: Writing NULL storage marker.\n", "[0.66s] Conversion complete.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "INFO:sentence_transformers.SentenceTransformer:Use pytorch device_name: mps\n", "INFO:sentence_transformers.SentenceTransformer:Load pretrained SentenceTransformer: facebook/contriever-msmarco\n", "WARNING:sentence_transformers.SentenceTransformer:No sentence-transformers model found with name facebook/contriever-msmarco. Creating a new one with mean pooling.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "✅ CSR conversion successful.\n", "INFO: Replaced original index with CSR-pruned version at 'knowledge.index'\n", "[read_HNSW - CSR NL v4] Reading metadata & CSR indices (manual offset)...\n", "INFO: Terminating session server process (PID: 70979)...\n", "[read_HNSW NL v4] Read levels vector, size: 5\n", "[read_HNSW NL v4] Reading Compact Storage format indices...\n", "[read_HNSW NL v4] Read compact_level_ptr, size: 10\n", "[read_HNSW NL v4] Read compact_node_offsets, size: 6\n", "[read_HNSW NL v4] Read entry_point: 4, max_level: 0\n", "[read_HNSW NL v4] Read storage fourcc: 0x6c6c756e\n", "[read_HNSW NL v4 FIX] Detected FileIOReader. Neighbors size field offset: 326\n", "[read_HNSW NL v4] Reading neighbors data into memory.\n", "[read_HNSW NL v4] Read neighbors data, size: 20\n", "[read_HNSW NL v4] Finished reading metadata and CSR indices.\n", "INFO: Skipping external storage loading, since is_recompute is true.\n", "INFO: Server process terminated.\n", "🔍 DEBUG LeannSearcher.search() called:\n", " Query: 'programming languages'\n", " Top_k: 2\n", " Additional kwargs: {'recompute_beighbor_embeddings': True}\n", "INFO: Computing embeddings for 1 chunks using SentenceTransformer model 'facebook/contriever-msmarco'...\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "Batches: 100%|██████████| 1/1 [00:00<00:00, 12.04it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Generated embedding shape: (1, 768)\n", " Embedding time: 1.2403802871704102 seconds\n", "INFO: Starting session-level embedding server for 'leann_backend_hnsw.hnsw_embedding_server'...\n", "INFO: Running command from project root: /Users/yichuan/Desktop/code/LEANN/leann\n", "INFO: Command: /Users/yichuan/Desktop/code/LEANN/leann/.venv/bin/python -m leann_backend_hnsw.hnsw_embedding_server --zmq-port 5557 --model-name facebook/contriever-msmarco --passages-file knowledge.leann.meta.json --disable-warmup\n", "INFO: Server process started with PID: 71209\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\n", "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n", "To disable this warning, you can either:\n", "\t- Avoid using `tokenizers` before the fork if possible\n", "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "✅ Embedding server is up and ready for this session.\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: INFO: Starting backend auto-discovery...\n", "ZmqDistanceComputer initialized: d=768, metric=0\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: INFO: Registering backend 'diskann'\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: INFO: Backend auto-discovery finished.\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: INFO: Registering backend 'hnsw'\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Loading tokenizer for facebook/contriever-msmarco...\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Tokenizer loaded successfully!\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: MPS available: True\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: CUDA available: False\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Using MPS device (Apple Silicon)\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Starting HNSW server on port 5557 with model facebook/contriever-msmarco\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Loading model facebook/contriever-msmarco... (this may take a while if downloading)\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Model facebook/contriever-msmarco loaded successfully!\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Loaded label map with 5 entries\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Initialized lazy passage loading for 5 passages\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Warmup disabled or no passages available (enable_warmup=False, passages=5)\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: HNSW ZMQ server listening on port 5557\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Started HNSW ZMQ server thread on port 5557\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Received ZMQ request of size 3 bytes\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload length: 1\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[0]: - [4]\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Request for 1 node embeddings\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Total batch size: 1, max_batch_size: 128\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG zmq_server_thread: Final 'hidden' array | Shape: (1, 768) | Dtype: float32 | Has NaN/Inf: False\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Serialize time: 0.000154 seconds\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: ZMQ E2E time: 0.131017 seconds\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Received ZMQ request of size 3849 bytes\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload length: 2\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[0]: - [0, 1, 2, 3]\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[1]: - [0.028167724609375, -0.01134490966796875, 0.044586181640625, -0.017486572265625, -0.028564453125, -0...\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Distance calculation request received\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Node IDs: [0, 1, 2, 3]\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Query vector dim: 768\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Passages loaded: 5\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 0\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 0, length: 37\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 1\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 1, length: 64\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 2\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 2, length: 38\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 3\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 3, length: 36\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Sending distance response with 4 distances\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Distance calculation E2E time: 0.131733 seconds\n", " Search time: 4.379124879837036 seconds\n", " Backend returned: labels=2 results\n", " Processing 2 passage IDs:\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "INFO:leann.chat:Attempting to create LLM of type='hf' with model='Qwen/Qwen3-0.6B'\n", "INFO:leann.chat:Initializing HFChat with model='Qwen/Qwen3-0.6B'\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " 1. passage_id='8f6d8742-3659-4d2f-ac45-377fd69b031e' -> SUCCESS: C# is a powerful programming language...\n", " 2. passage_id='837f1f70-3c8c-498f-867d-06a063aa2a6e' -> SUCCESS: Python is a powerful programming language and it is very popular...\n", " Final enriched results: 2 passages\n", "[SearchResult(id='8f6d8742-3659-4d2f-ac45-377fd69b031e', score=np.float32(1.4450607), text='C# is a powerful programming language', metadata={}), SearchResult(id='837f1f70-3c8c-498f-867d-06a063aa2a6e', score=np.float32(1.394449), text='Python is a powerful programming language and it is very popular', metadata={})]\n", "[read_HNSW - CSR NL v4] Reading metadata & CSR indices (manual offset)...\n", "[read_HNSW NL v4] Read levels vector, size: 5\n", "[read_HNSW NL v4] Reading Compact Storage format indices...\n", "[read_HNSW NL v4] Read compact_level_ptr, size: 10\n", "[read_HNSW NL v4] Read compact_node_offsets, size: 6\n", "[read_HNSW NL v4] Read entry_point: 4, max_level: 0\n", "[read_HNSW NL v4] Read storage fourcc: 0x6c6c756e\n", "[read_HNSW NL v4 FIX] Detected FileIOReader. Neighbors size field offset: 326\n", "[read_HNSW NL v4] Reading neighbors data into memory.\n", "[read_HNSW NL v4] Read neighbors data, size: 20\n", "[read_HNSW NL v4] Finished reading metadata and CSR indices.\n", "INFO: Skipping external storage loading, since is_recompute is true.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "INFO:leann.chat:MPS is available. Using Apple Silicon GPU.\n", "Device set to use mps\n", "INFO:sentence_transformers.SentenceTransformer:Use pytorch device_name: mps\n", "INFO:sentence_transformers.SentenceTransformer:Load pretrained SentenceTransformer: facebook/contriever-msmarco\n", "WARNING:sentence_transformers.SentenceTransformer:No sentence-transformers model found with name facebook/contriever-msmarco. Creating a new one with mean pooling.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "🔍 DEBUG LeannSearcher.search() called:\n", " Query: 'Compare the two retrieved programming languages and say which one is more popular today. Respond in a single well-formed sentence.'\n", " Top_k: 2\n", " Additional kwargs: {'recompute_beighbor_embeddings': True}\n", "INFO: Computing embeddings for 1 chunks using SentenceTransformer model 'facebook/contriever-msmarco'...\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "Batches: 100%|██████████| 1/1 [00:00<00:00, 7.66it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Generated embedding shape: (1, 768)\n", " Embedding time: 1.5981061458587646 seconds\n", "INFO: Port 5557 is in use. Checking server compatibility...\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Received ZMQ request of size 17 bytes\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload length: 1\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[0]: - __QUERY_MODEL__\n", "✅ Existing server on port 5557 is using correct model: facebook/contriever-msmarco\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Received ZMQ request of size 21 bytes\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload length: 1\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[0]: - __QUERY_META_PATH__\n", "✅ Existing server on port 5557 is using correct meta path: knowledge.leann.meta.json\n", "✅ Server on port 5557 is compatible and ready to use.\n", "ZmqDistanceComputer initialized: d=768, metric=0\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Received ZMQ request of size 3 bytes\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload length: 1\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[0]: - [4]\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Request for 1 node embeddings\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Total batch size: 1, max_batch_size: 128\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG zmq_server_thread: Final 'hidden' array | Shape: (1, 768) | Dtype: float32 | Has NaN/Inf: False\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Serialize time: 0.000330 seconds\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: ZMQ E2E time: 0.165497 seconds\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Received ZMQ request of size 3849 bytes\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload length: 2\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[0]: - [0, 1, 2, 3]\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: request_payload[1]: - [0.0753173828125, -0.00579071044921875, 0.0662841796875, 0.014923095703125, -0.0064544677734375, -0....\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Distance calculation request received\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Node IDs: [0, 1, 2, 3]\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Query vector dim: 768\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Passages loaded: 5\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 0\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 0, length: 37\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 1\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 1, length: 64\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 2\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 2, length: 38\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Looking up passage ID 3\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: DEBUG: Found text for ID 3, length: 36\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Sending distance response with 4 distances\n", "[leann_backend_hnsw.hnsw_embedding_server LOG]: Distance calculation E2E time: 0.082911 seconds\n", " Search time: 0.2542300224304199 seconds\n", " Backend returned: labels=2 results\n", " Processing 2 passage IDs:\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "INFO:leann.chat:Generating text with Hugging Face model with params: {'max_length': 500, 'num_return_sequences': 1}\n", "Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`.\n", "Both `max_new_tokens` (=256) and `max_length`(=500) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " 1. passage_id='837f1f70-3c8c-498f-867d-06a063aa2a6e' -> SUCCESS: Python is a powerful programming language and it is very popular...\n", " 2. passage_id='8f6d8742-3659-4d2f-ac45-377fd69b031e' -> SUCCESS: C# is a powerful programming language...\n", " Final enriched results: 2 passages\n", "Also, make sure you answer in a single sentence.\n", "Answer:\n", "The retrieved context says that Python and C# are both powerful programming languages, but it does not provide any specific information about their popularity. However, based on general knowledge, Python is more popular than C# in current markets.\n", "The answer should be in a single sentence.\n", "Answer:\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "So the final answer is:\n", "\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "Answer:\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "So the final answer is:\n", "\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "Answer:\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "So the final answer is:\n", "\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "Answer:\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "So the final answer is:\n", "\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "Answer:\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "So the final answer is:\n", "\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "Answer:\n", "Python is a more popular programming language than C# today.\n", "---\n", "\n", "So the final answer\n" ] } ], "source": [ "from leann.api import LeannBuilder, LeannSearcher, LeannChat\n", "# 1. Build index (no embeddings stored!)\n", "builder = LeannBuilder(backend_name=\"hnsw\")\n", "builder.add_text(\"C# is a powerful programming language\")\n", "builder.add_text(\"Python is a powerful programming language and it is very popular\")\n", "builder.add_text(\"Machine learning transforms industries\") \n", "builder.add_text(\"Neural networks process complex data\")\n", "builder.add_text(\"Leann is a great storage saving engine for RAG on your macbook\")\n", "builder.build_index(\"knowledge.leann\")\n", "# 2. Search with real-time embeddings\n", "searcher = LeannSearcher(\"knowledge.leann\")\n", "results = searcher.search(\"programming languages\", top_k=2, recompute_beighbor_embeddings=True)\n", "print(results)\n", "\n", "llm_config = {\"type\": \"hf\", \"model\": \"Qwen/Qwen3-0.6B\"}\n", "\n", "chat = LeannChat(index_path=\"knowledge.leann\", llm_config=llm_config)\n", "\n", "response = chat.ask(\n", " \"Compare the two retrieved programming languages and say which one is more popular today. Respond in a single well-formed sentence.\",\n", " top_k=2,\n", " recompute_beighbor_embeddings=True,\n", ")\n", "print(response)" ] } ], "metadata": { "kernelspec": { "display_name": ".venv", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.12" } }, "nbformat": 4, "nbformat_minor": 2 }