docs

* feat: Add GitHub PR and issue templates for better contributor experience

* simplify: Make templates more concise and user-friendly

* fix: enable is_compact=False, is_recompute=True

* feat: update when recompute

* test

* fix: real recompute

* refactor

* fix: compare with no-recompute

* fix: test

.gitignore

@@ -101,4 +101,3 @@ CLAUDE.local.md
 .claude/*.local.*
 .claude/local/*
 benchmarks/data/
-test_add/*

apps/base_rag_example.py

@@ -11,6 +11,7 @@ from typing import Any
 import dotenv
 from leann.api import LeannBuilder, LeannChat
 from leann.registry import register_project_directory
+from leann.settings import resolve_ollama_host, resolve_openai_api_key, resolve_openai_base_url
 
 dotenv.load_dotenv()
 
@@ -78,6 +79,24 @@ class BaseRAGExample(ABC):
             choices=["sentence-transformers", "openai", "mlx", "ollama"],
             help="Embedding backend mode (default: sentence-transformers), we provide sentence-transformers, openai, mlx, or ollama",
         )
+        embedding_group.add_argument(
+            "--embedding-host",
+            type=str,
+            default=None,
+            help="Override Ollama-compatible embedding host",
+        )
+        embedding_group.add_argument(
+            "--embedding-api-base",
+            type=str,
+            default=None,
+            help="Base URL for OpenAI-compatible embedding services",
+        )
+        embedding_group.add_argument(
+            "--embedding-api-key",
+            type=str,
+            default=None,
+            help="API key for embedding service (defaults to OPENAI_API_KEY)",
+        )
 
         # LLM parameters
         llm_group = parser.add_argument_group("LLM Parameters")
@@ -97,8 +116,8 @@ class BaseRAGExample(ABC):
         llm_group.add_argument(
             "--llm-host",
             type=str,
-            default="http://localhost:11434",
-            help="Host for Ollama API (default: http://localhost:11434)",
+            default=None,
+            help="Host for Ollama-compatible APIs (defaults to LEANN_OLLAMA_HOST/OLLAMA_HOST)",
         )
         llm_group.add_argument(
             "--thinking-budget",
@@ -107,6 +126,18 @@ class BaseRAGExample(ABC):
             default=None,
             help="Thinking budget for reasoning models (low/medium/high). Supported by GPT-Oss:20b and other reasoning models.",
         )
+        llm_group.add_argument(
+            "--llm-api-base",
+            type=str,
+            default=None,
+            help="Base URL for OpenAI-compatible APIs",
+        )
+        llm_group.add_argument(
+            "--llm-api-key",
+            type=str,
+            default=None,
+            help="API key for OpenAI-compatible APIs (defaults to OPENAI_API_KEY)",
+        )
 
         # AST Chunking parameters
         ast_group = parser.add_argument_group("AST Chunking Parameters")
@@ -205,9 +236,13 @@ class BaseRAGExample(ABC):
 
         if args.llm == "openai":
             config["model"] = args.llm_model or "gpt-4o"
+            config["base_url"] = resolve_openai_base_url(args.llm_api_base)
+            resolved_key = resolve_openai_api_key(args.llm_api_key)
+            if resolved_key:
+                config["api_key"] = resolved_key
         elif args.llm == "ollama":
             config["model"] = args.llm_model or "llama3.2:1b"
-            config["host"] = args.llm_host
+            config["host"] = resolve_ollama_host(args.llm_host)
         elif args.llm == "hf":
             config["model"] = args.llm_model or "Qwen/Qwen2.5-1.5B-Instruct"
         elif args.llm == "simulated":
@@ -223,10 +258,20 @@ class BaseRAGExample(ABC):
         print(f"\n[Building Index] Creating {self.name} index...")
         print(f"Total text chunks: {len(texts)}")
 
+        embedding_options: dict[str, Any] = {}
+        if args.embedding_mode == "ollama":
+            embedding_options["host"] = resolve_ollama_host(args.embedding_host)
+        elif args.embedding_mode == "openai":
+            embedding_options["base_url"] = resolve_openai_base_url(args.embedding_api_base)
+            resolved_embedding_key = resolve_openai_api_key(args.embedding_api_key)
+            if resolved_embedding_key:
+                embedding_options["api_key"] = resolved_embedding_key
+
         builder = LeannBuilder(
             backend_name=args.backend_name,
             embedding_model=args.embedding_model,
             embedding_mode=args.embedding_mode,
+            embedding_options=embedding_options or None,
             graph_degree=args.graph_degree,
             complexity=args.build_complexity,
             is_compact=not args.no_compact,

docs/configuration-guide.md

@@ -83,6 +83,81 @@ ollama pull nomic-embed-text
 
 </details>
 
+## Local & Remote Inference Endpoints
+
+> Applies to both LLMs (`leann ask`) and embeddings (`leann build`).
+
+LEANN now treats Ollama, LM Studio, and other OpenAI-compatible runtimes as first-class providers. You can point LEANN at any compatible endpoint – either on the same machine or across the network – with a couple of flags or environment variables.
+
+### One-Time Environment Setup
+
+```bash
+# Works for OpenAI-compatible runtimes such as LM Studio, vLLM, SGLang, llamafile, etc.
+export OPENAI_API_KEY="your-key"            # or leave unset for local servers that do not check keys
+export OPENAI_BASE_URL="http://localhost:1234/v1"
+
+# Ollama-compatible runtimes (Ollama, Ollama on another host, llamacpp-server, etc.)
+export LEANN_OLLAMA_HOST="http://localhost:11434"   # falls back to OLLAMA_HOST or LOCAL_LLM_ENDPOINT
+```
+
+LEANN also recognises `LEANN_LOCAL_LLM_HOST` (highest priority), `LEANN_OPENAI_BASE_URL`, and `LOCAL_OPENAI_BASE_URL`, so existing scripts continue to work.
+
+### Passing Hosts Per Command
+
+```bash
+# Build an index with a remote embedding server
+leann build my-notes \
+  --docs ./notes \
+  --embedding-mode openai \
+  --embedding-model text-embedding-qwen3-embedding-0.6b \
+  --embedding-api-base http://192.168.1.50:1234/v1 \
+  --embedding-api-key local-dev-key
+
+# Query using a local LM Studio instance via OpenAI-compatible API
+leann ask my-notes \
+  --llm openai \
+  --llm-model qwen3-8b \
+  --api-base http://localhost:1234/v1 \
+  --api-key local-dev-key
+
+# Query an Ollama instance running on another box
+leann ask my-notes \
+  --llm ollama \
+  --llm-model qwen3:14b \
+  --host http://192.168.1.101:11434
+```
+
+⚠️ **Make sure the endpoint is reachable**: when your inference server runs on a home/workstation and the index/search job runs in the cloud, the server must be able to reach the host you configured. Typical options include:
+
+- Expose a public IP (and open the relevant port) on the machine that hosts LM Studio/Ollama.
+- Configure router or cloud provider port forwarding.
+- Tunnel traffic through tools like `tailscale`, `cloudflared`, or `ssh -R`.
+
+When you set these options while building an index, LEANN stores them in `meta.json`. Any subsequent `leann ask` or searcher process automatically reuses the same provider settings – even when we spawn background embedding servers. This makes the “server without GPU talking to my local workstation” workflow from [issue #80](https://github.com/yichuan-w/LEANN/issues/80#issuecomment-2287230548) work out-of-the-box.
+
+**Tip:** If your runtime does not require an API key (many local stacks don’t), leave `--api-key` unset. LEANN will skip injecting credentials.
+
+### Python API Usage
+
+You can pass the same configuration from Python:
+
+```python
+from leann.api import LeannBuilder
+
+builder = LeannBuilder(
+    backend_name="hnsw",
+    embedding_mode="openai",
+    embedding_model="text-embedding-qwen3-embedding-0.6b",
+    embedding_options={
+        "base_url": "http://192.168.1.50:1234/v1",
+        "api_key": "local-dev-key",
+    },
+)
+builder.build_index("./indexes/my-notes", chunks)
+```
+
+`embedding_options` is persisted to the index `meta.json`, so subsequent `LeannSearcher` or `LeannChat` sessions automatically reuse the same provider settings (the embedding server manager forwards them to the provider for you).
+
 ## Index Selection: Matching Your Scale
 
 ### HNSW (Hierarchical Navigable Small World)

examples/dynamic_add_leann_no_recompute.py

@@ -1,380 +0,0 @@
-"""
-Dynamic add example for LEANN using HNSW backend without recompute.
-
-- Builds a base index from a directory of documents
-- Incrementally adds new documents without recomputing stored embeddings
-
-Defaults:
-- Base data: /Users/yichuan/Desktop/code/LEANN/leann/data
-- Incremental data: /Users/yichuan/Desktop/code/LEANN/leann/test_add
-- Index path: <index_dir>/documents.leann
-
-Usage examples:
-  uv run python examples/dynamic_add_leann_no_recompute.py --build-base \
-    --base-dir /Users/yichuan/Desktop/code/LEANN/leann/data \
-    --index-dir ./test_doc_files
-
-  uv run python examples/dynamic_add_leann_no_recompute.py --add-incremental \
-    --add-dir /Users/yichuan/Desktop/code/LEANN/leann/test_add \
-    --index-dir ./test_doc_files
-
-Quick recompute test (both true):
-  # Recompute build
-  uv run python examples/dynamic_add_leann_no_recompute.py --build-base \
-    --recompute-build --ef-construction 200 \
-    --base-dir /Users/yichuan/Desktop/code/LEANN/leann/data \
-    --index-dir ./test_doc_files --index-name documents.leann
-
-  # Recompute add
-  uv run python examples/dynamic_add_leann_no_recompute.py --add-incremental \
-    --recompute-add --ef-construction 32 \
-    --add-dir /Users/yichuan/Desktop/code/LEANN/leann/test_add \
-    --index-dir ./test_doc_files --index-name documents.leann
-"""
-
-import argparse
-import json
-import pickle
-import sys
-from pathlib import Path
-from typing import Any, Optional
-
-# Ensure we can import from the local packages and apps folders
-ROOT = Path(__file__).resolve().parents[1]
-CORE_SRC = ROOT / "packages" / "leann-core" / "src"
-HNSW_PKG_DIR = ROOT / "packages" / "leann-backend-hnsw"
-APPS_DIR = ROOT / "apps"
-
-
-# Prefer the installed backend if available (it contains the compiled extension)
-def _prefer_installed(pkg_name: str) -> bool:
-    try:
-        import importlib
-        import importlib.util
-
-        spec = importlib.util.find_spec(pkg_name)
-        if spec and spec.origin and "site-packages" in spec.origin:
-            # ensure the faiss shim/extension is importable from the installed package
-            importlib.import_module(f"{pkg_name}.faiss")
-            return True
-    except Exception:
-        pass
-    return False
-
-
-# Prepend paths, but only add the repo backend if the installed one is not present
-paths_to_prepend = [CORE_SRC, APPS_DIR]
-if not _prefer_installed("leann_backend_hnsw"):
-    paths_to_prepend.insert(1, HNSW_PKG_DIR)
-
-for p in paths_to_prepend:
-    p_str = str(p)
-    if p_str not in sys.path:
-        sys.path.insert(0, p_str)
-
-# Defer non-stdlib imports until after sys.path setup within functions (avoid E402)
-
-
-def _load_documents(data_dir: str, required_exts: Optional[list[str]] = None) -> list[Any]:
-    from llama_index.core import SimpleDirectoryReader  # type: ignore
-
-    reader_kwargs: dict[str, Any] = {"recursive": True, "encoding": "utf-8"}
-    if required_exts:
-        reader_kwargs["required_exts"] = required_exts
-    documents = SimpleDirectoryReader(data_dir, **reader_kwargs).load_data(show_progress=True)
-    return documents
-
-
-def _ensure_index_dir(index_dir: Path) -> None:
-    index_dir.mkdir(parents=True, exist_ok=True)
-
-
-def _index_files(index_path: Path) -> tuple[Path, Path, Path]:
-    """Return (passages.jsonl, passages.idx, index.index) paths for a given index base path.
-
-    Note: HNSWBackend writes the FAISS index using the stem (without .leann),
-    i.e., for base 'documents.leann' the file is 'documents.index'. We prefer the
-    existing file among candidates.
-    """
-    passages_file = index_path.parent / f"{index_path.name}.passages.jsonl"
-    offsets_file = index_path.parent / f"{index_path.name}.passages.idx"
-    candidate_name_index = index_path.parent / f"{index_path.name}.index"
-    candidate_stem_index = index_path.parent / f"{index_path.stem}.index"
-    index_file = candidate_stem_index if candidate_stem_index.exists() else candidate_name_index
-    return passages_file, offsets_file, index_file
-
-
-def _read_meta(index_path: Path) -> dict[str, Any]:
-    meta_path = index_path.parent / f"{index_path.name}.meta.json"
-    if not meta_path.exists():
-        raise FileNotFoundError(f"Metadata file not found: {meta_path}")
-    with open(meta_path, encoding="utf-8") as f:
-        return json.load(f)
-
-
-def _autodetect_index_base(index_dir: Path) -> Optional[Path]:
-    """If exactly one *.leann.meta.json exists, return its base path (without .meta.json)."""
-    candidates = list(index_dir.glob("*.leann.meta.json"))
-    if len(candidates) == 1:
-        meta = candidates[0]
-        base = meta.with_suffix("")  # remove .json
-        base = base.with_suffix("")  # remove .meta
-        return base
-    return None
-
-
-def _load_offset_map(offsets_file: Path) -> dict[str, int]:
-    if not offsets_file.exists():
-        return {}
-    with open(offsets_file, "rb") as f:
-        return pickle.load(f)
-
-
-def _next_numeric_id(existing_ids: list[str]) -> int:
-    numeric_ids = [int(x) for x in existing_ids if x.isdigit()]
-    if not numeric_ids:
-        return 0
-    return max(numeric_ids) + 1
-
-
-def build_base_index(
-    base_dir: str,
-    index_dir: str,
-    index_name: str,
-    embedding_model: str,
-    embedding_mode: str,
-    chunk_size: int,
-    chunk_overlap: int,
-    file_types: Optional[list[str]] = None,
-    max_items: int = -1,
-    ef_construction: Optional[int] = None,
-    recompute_build: bool = False,
-) -> str:
-    print(f"Building base index from: {base_dir}")
-    documents = _load_documents(base_dir, required_exts=file_types)
-    if not documents:
-        raise ValueError(f"No documents found in base_dir: {base_dir}")
-
-    from chunking import create_text_chunks
-
-    texts = create_text_chunks(
-        documents,
-        chunk_size=chunk_size,
-        chunk_overlap=chunk_overlap,
-        use_ast_chunking=False,
-    )
-    if max_items > 0 and len(texts) > max_items:
-        texts = texts[:max_items]
-        print(f"Limiting to {max_items} chunks")
-
-    index_dir_path = Path(index_dir)
-    _ensure_index_dir(index_dir_path)
-    index_path = index_dir_path / index_name
-
-    print("Creating HNSW index (non-compact)...")
-    from leann.api import LeannBuilder
-    from leann.registry import register_project_directory
-
-    builder = LeannBuilder(
-        backend_name="hnsw",
-        embedding_model=embedding_model,
-        embedding_mode=embedding_mode,
-        is_recompute=recompute_build,
-        is_compact=False,
-        efConstruction=(ef_construction if ef_construction is not None else 200),
-    )
-    for t in texts:
-        builder.add_text(t)
-    builder.build_index(str(index_path))
-
-    # Register for discovery
-    register_project_directory(Path.cwd())
-
-    print(f"Base index built at: {index_path}")
-    return str(index_path)
-
-
-def add_incremental(
-    add_dir: str,
-    index_dir: str,
-    index_name: Optional[str] = None,
-    embedding_model: Optional[str] = None,
-    embedding_mode: Optional[str] = None,
-    chunk_size: int = 256,
-    chunk_overlap: int = 128,
-    file_types: Optional[list[str]] = None,
-    max_items: int = -1,
-    ef_construction: Optional[int] = None,
-    recompute_add: bool = False,
-) -> str:
-    print(f"Adding incremental data from: {add_dir}")
-    index_dir_path = Path(index_dir)
-    index_path = index_dir_path / (index_name or "documents.leann")
-
-    # If specified base doesn't exist, try to auto-detect an existing base
-    try:
-        _read_meta(index_path)
-    except FileNotFoundError:
-        auto_base = _autodetect_index_base(index_dir_path)
-        if auto_base is not None:
-            print(f"Auto-detected index base: {auto_base.name}")
-            index_path = auto_base
-            _read_meta(index_path)
-        else:
-            raise FileNotFoundError(
-                f"No index metadata found for base '{index_path.name}'. Build base first with --build-base "
-                f"or provide --index-name to match an existing index (e.g., 'test_doc_files.leann')."
-            )
-
-    # Prepare validated context from core (checks backend/no-recompute and resolves embedding defaults)
-    from leann.api import create_incremental_add_context, incremental_add_texts_with_context
-
-    ctx = create_incremental_add_context(
-        str(index_path),
-        embedding_model=embedding_model,
-        embedding_mode=embedding_mode,
-        data_dir=add_dir,
-        required_exts=file_types,
-        chunk_size=chunk_size,
-        chunk_overlap=chunk_overlap,
-        max_items=max_items,
-    )
-
-    # Use prepared texts from context to perform the add
-    prepared_texts = ctx.prepared_texts or []
-    if not prepared_texts:
-        print("No new chunks to add.")
-        return str(index_path)
-
-    added = incremental_add_texts_with_context(
-        ctx,
-        prepared_texts,
-        ef_construction=ef_construction,
-        recompute=recompute_add,
-    )
-
-    print(f"Incremental add completed. Added {added} chunks. Index: {index_path}")
-    return str(index_path)
-
-
-def main():
-    parser = argparse.ArgumentParser(
-        description="Dynamic add to LEANN HNSW index without recompute",
-        formatter_class=argparse.RawDescriptionHelpFormatter,
-    )
-
-    parser.add_argument("--build-base", action="store_true", help="Build base index")
-    parser.add_argument("--add-incremental", action="store_true", help="Add incremental data")
-
-    parser.add_argument(
-        "--base-dir",
-        type=str,
-        default="/Users/yichuan/Desktop/code/LEANN/leann/data",
-        help="Base data directory",
-    )
-    parser.add_argument(
-        "--add-dir",
-        type=str,
-        default="/Users/yichuan/Desktop/code/LEANN/leann/test_add",
-        help="Incremental data directory",
-    )
-    parser.add_argument(
-        "--index-dir",
-        type=str,
-        default="./test_doc_files",
-        help="Directory containing the index",
-    )
-    parser.add_argument(
-        "--index-name",
-        type=str,
-        default="documents.leann",
-        help=(
-            "Index base file name. If you built via document_rag.py, use 'test_doc_files.leann'. "
-            "Default: documents.leann"
-        ),
-    )
-
-    parser.add_argument(
-        "--embedding-model",
-        type=str,
-        default="facebook/contriever",
-        help="Embedding model name",
-    )
-    parser.add_argument(
-        "--embedding-mode",
-        type=str,
-        default="sentence-transformers",
-        choices=["sentence-transformers", "openai", "mlx", "ollama"],
-        help="Embedding backend mode",
-    )
-
-    parser.add_argument("--chunk-size", type=int, default=256)
-    parser.add_argument("--chunk-overlap", type=int, default=128)
-    parser.add_argument("--file-types", nargs="+", default=None)
-    parser.add_argument("--max-items", type=int, default=-1)
-    parser.add_argument("--ef-construction", type=int, default=32)
-    parser.add_argument(
-        "--recompute-add", action="store_true", help="Enable recompute-mode add (non-compact only)"
-    )
-    parser.add_argument(
-        "--recompute-build",
-        action="store_true",
-        help="Enable recompute-mode base build (non-compact only)",
-    )
-
-    args = parser.parse_args()
-
-    if not args.build_base and not args.add_incremental:
-        print("Nothing to do. Use --build-base and/or --add-incremental.")
-        return
-
-    index_path_str: Optional[str] = None
-
-    if args.build_base:
-        index_path_str = build_base_index(
-            base_dir=args.base_dir,
-            index_dir=args.index_dir,
-            index_name=args.index_name,
-            embedding_model=args.embedding_model,
-            embedding_mode=args.embedding_mode,
-            chunk_size=args.chunk_size,
-            chunk_overlap=args.chunk_overlap,
-            file_types=args.file_types,
-            max_items=args.max_items,
-            ef_construction=args.ef_construction,
-            recompute_build=args.recompute_build,
-        )
-
-    if args.add_incremental:
-        index_path_str = add_incremental(
-            add_dir=args.add_dir,
-            index_dir=args.index_dir,
-            index_name=args.index_name,
-            embedding_model=args.embedding_model,
-            embedding_mode=args.embedding_mode,
-            chunk_size=args.chunk_size,
-            chunk_overlap=args.chunk_overlap,
-            file_types=args.file_types,
-            max_items=args.max_items,
-            ef_construction=args.ef_construction,
-            recompute_add=args.recompute_add,
-        )
-
-    # Optional: quick test query using searcher
-    if index_path_str:
-        try:
-            from leann.api import LeannSearcher
-
-            searcher = LeannSearcher(index_path_str)
-            query = "what is LEANN?"
-            if args.add_incremental:
-                query = "what is the multi vector search and how it works?"
-            results = searcher.search(query, top_k=5)
-            if results:
-                print(f"Sample result: {results[0].text[:80]}...")
-        except Exception:
-            pass
-
-
-if __name__ == "__main__":
-    main()

examples/dynamic_update_no_recompute.py

@@ -0,0 +1,404 @@
+"""Dynamic HNSW update demo without compact storage.
+
+This script reproduces the minimal scenario we used while debugging on-the-fly
+recompute:
+
+1. Build a non-compact HNSW index from the first few paragraphs of a text file.
+2. Print the top results with `recompute_embeddings=True`.
+3. Append additional paragraphs with :meth:`LeannBuilder.update_index`.
+4. Run the same query again to show the newly inserted passages.
+
+Run it with ``uv`` (optionally pointing LEANN_HNSW_LOG_PATH at a file to inspect
+ZMQ activity)::
+
+    LEANN_HNSW_LOG_PATH=embedding_fetch.log \
+    uv run -m examples.dynamic_update_no_recompute \
+      --index-path .leann/examples/leann-demo.leann
+
+By default the script builds an index from ``data/2501.14312v1 (1).pdf`` and
+then updates it with LEANN-related material from ``data/2506.08276v1.pdf``.
+It issues the query "What's LEANN?" before and after the update to show how the
+new passages become immediately searchable. The script uses the
+``sentence-transformers/all-MiniLM-L6-v2`` model with ``is_recompute=True`` so
+Faiss pulls existing vectors on demand via the ZMQ embedding server, while
+freshly added passages are embedded locally just like the initial build.
+
+To make storage comparisons easy, the script can also build a matching
+``is_recompute=False`` baseline (enabled by default) and report the index size
+delta after the update. Disable the baseline run with
+``--skip-compare-no-recompute`` if you only need the recompute flow.
+"""
+
+import argparse
+import json
+from collections.abc import Iterable
+from pathlib import Path
+from typing import Any
+
+from leann.api import LeannBuilder, LeannSearcher
+from leann.registry import register_project_directory
+
+from apps.chunking import create_text_chunks
+
+REPO_ROOT = Path(__file__).resolve().parents[1]
+
+DEFAULT_QUERY = "What's LEANN?"
+DEFAULT_INITIAL_FILES = [REPO_ROOT / "data" / "2501.14312v1 (1).pdf"]
+DEFAULT_UPDATE_FILES = [REPO_ROOT / "data" / "2506.08276v1.pdf"]
+
+
+def load_chunks_from_files(paths: list[Path]) -> list[str]:
+    from llama_index.core import SimpleDirectoryReader
+
+    documents = []
+    for path in paths:
+        p = path.expanduser().resolve()
+        if not p.exists():
+            raise FileNotFoundError(f"Input path not found: {p}")
+        if p.is_dir():
+            reader = SimpleDirectoryReader(str(p), recursive=False)
+            documents.extend(reader.load_data(show_progress=True))
+        else:
+            reader = SimpleDirectoryReader(input_files=[str(p)])
+            documents.extend(reader.load_data(show_progress=True))
+
+    if not documents:
+        return []
+
+    chunks = create_text_chunks(
+        documents,
+        chunk_size=512,
+        chunk_overlap=128,
+        use_ast_chunking=False,
+    )
+    return [c for c in chunks if isinstance(c, str) and c.strip()]
+
+
+def run_search(index_path: Path, query: str, top_k: int, *, recompute_embeddings: bool) -> list:
+    searcher = LeannSearcher(str(index_path))
+    try:
+        return searcher.search(
+            query=query,
+            top_k=top_k,
+            recompute_embeddings=recompute_embeddings,
+            batch_size=16,
+        )
+    finally:
+        searcher.cleanup()
+
+
+def print_results(title: str, results: Iterable) -> None:
+    print(f"\n=== {title} ===")
+    res_list = list(results)
+    print(f"results count: {len(res_list)}")
+    print("passages:")
+    if not res_list:
+        print("  (no passages returned)")
+    for res in res_list:
+        snippet = res.text.replace("\n", " ")[:120]
+        print(f"  - {res.id}: {snippet}... (score={res.score:.4f})")
+
+
+def build_initial_index(
+    index_path: Path,
+    paragraphs: list[str],
+    model_name: str,
+    embedding_mode: str,
+    is_recompute: bool,
+) -> None:
+    builder = LeannBuilder(
+        backend_name="hnsw",
+        embedding_model=model_name,
+        embedding_mode=embedding_mode,
+        is_compact=False,
+        is_recompute=is_recompute,
+    )
+    for idx, passage in enumerate(paragraphs):
+        builder.add_text(passage, metadata={"id": str(idx)})
+    builder.build_index(str(index_path))
+
+
+def update_index(
+    index_path: Path,
+    start_id: int,
+    paragraphs: list[str],
+    model_name: str,
+    embedding_mode: str,
+    is_recompute: bool,
+) -> None:
+    updater = LeannBuilder(
+        backend_name="hnsw",
+        embedding_model=model_name,
+        embedding_mode=embedding_mode,
+        is_compact=False,
+        is_recompute=is_recompute,
+    )
+    for offset, passage in enumerate(paragraphs, start=start_id):
+        updater.add_text(passage, metadata={"id": str(offset)})
+    updater.update_index(str(index_path))
+
+
+def ensure_index_dir(index_path: Path) -> None:
+    index_path.parent.mkdir(parents=True, exist_ok=True)
+
+
+def cleanup_index_files(index_path: Path) -> None:
+    """Remove leftover index artifacts for a clean rebuild."""
+
+    parent = index_path.parent
+    if not parent.exists():
+        return
+    stem = index_path.stem
+    for file in parent.glob(f"{stem}*"):
+        if file.is_file():
+            file.unlink()
+
+
+def index_file_size(index_path: Path) -> int:
+    """Return the size of the primary .index file for the given index path."""
+
+    index_file = index_path.parent / f"{index_path.stem}.index"
+    return index_file.stat().st_size if index_file.exists() else 0
+
+
+def load_metadata_snapshot(index_path: Path) -> dict[str, Any] | None:
+    meta_path = index_path.parent / f"{index_path.name}.meta.json"
+    if not meta_path.exists():
+        return None
+    try:
+        return json.loads(meta_path.read_text())
+    except json.JSONDecodeError:
+        return None
+
+
+def run_workflow(
+    *,
+    label: str,
+    index_path: Path,
+    initial_paragraphs: list[str],
+    update_paragraphs: list[str],
+    model_name: str,
+    embedding_mode: str,
+    is_recompute: bool,
+    query: str,
+    top_k: int,
+) -> dict[str, Any]:
+    prefix = f"[{label}] " if label else ""
+
+    ensure_index_dir(index_path)
+    cleanup_index_files(index_path)
+
+    print(f"{prefix}Building initial index...")
+    build_initial_index(
+        index_path,
+        initial_paragraphs,
+        model_name,
+        embedding_mode,
+        is_recompute=is_recompute,
+    )
+
+    initial_size = index_file_size(index_path)
+    before_results = run_search(
+        index_path,
+        query,
+        top_k,
+        recompute_embeddings=is_recompute,
+    )
+
+    print(f"\n{prefix}Updating index with additional passages...")
+    update_index(
+        index_path,
+        start_id=len(initial_paragraphs),
+        paragraphs=update_paragraphs,
+        model_name=model_name,
+        embedding_mode=embedding_mode,
+        is_recompute=is_recompute,
+    )
+
+    after_results = run_search(
+        index_path,
+        query,
+        top_k,
+        recompute_embeddings=is_recompute,
+    )
+    updated_size = index_file_size(index_path)
+
+    return {
+        "initial_size": initial_size,
+        "updated_size": updated_size,
+        "delta": updated_size - initial_size,
+        "before_results": before_results,
+        "after_results": after_results,
+        "metadata": load_metadata_snapshot(index_path),
+    }
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument(
+        "--initial-files",
+        type=Path,
+        nargs="+",
+        default=DEFAULT_INITIAL_FILES,
+        help="Initial document files (PDF/TXT) used to build the base index",
+    )
+    parser.add_argument(
+        "--index-path",
+        type=Path,
+        default=Path(".leann/examples/leann-demo.leann"),
+        help="Destination index path (default: .leann/examples/leann-demo.leann)",
+    )
+    parser.add_argument(
+        "--initial-count",
+        type=int,
+        default=8,
+        help="Number of chunks to use from the initial documents (default: 8)",
+    )
+    parser.add_argument(
+        "--update-files",
+        type=Path,
+        nargs="*",
+        default=DEFAULT_UPDATE_FILES,
+        help="Additional documents to add during update (PDF/TXT)",
+    )
+    parser.add_argument(
+        "--update-count",
+        type=int,
+        default=4,
+        help="Number of chunks to append from update documents (default: 4)",
+    )
+    parser.add_argument(
+        "--update-text",
+        type=str,
+        default=(
+            "LEANN (Lightweight Embedding ANN) is an indexing toolkit focused on "
+            "recompute-aware HNSW graphs, allowing embeddings to be regenerated "
+            "on demand to keep disk usage minimal."
+        ),
+        help="Fallback text to append if --update-files is omitted",
+    )
+    parser.add_argument(
+        "--top-k",
+        type=int,
+        default=4,
+        help="Number of results to show for each search (default: 4)",
+    )
+    parser.add_argument(
+        "--query",
+        type=str,
+        default=DEFAULT_QUERY,
+        help="Query to run before/after the update",
+    )
+    parser.add_argument(
+        "--embedding-model",
+        type=str,
+        default="sentence-transformers/all-MiniLM-L6-v2",
+        help="Embedding model name",
+    )
+    parser.add_argument(
+        "--embedding-mode",
+        type=str,
+        default="sentence-transformers",
+        choices=["sentence-transformers", "openai", "mlx", "ollama"],
+        help="Embedding backend mode",
+    )
+    parser.add_argument(
+        "--compare-no-recompute",
+        dest="compare_no_recompute",
+        action="store_true",
+        help="Also run a baseline with is_recompute=False and report its index growth.",
+    )
+    parser.add_argument(
+        "--skip-compare-no-recompute",
+        dest="compare_no_recompute",
+        action="store_false",
+        help="Skip building the no-recompute baseline.",
+    )
+    parser.set_defaults(compare_no_recompute=True)
+    args = parser.parse_args()
+
+    ensure_index_dir(args.index_path)
+    register_project_directory(REPO_ROOT)
+
+    initial_chunks = load_chunks_from_files(list(args.initial_files))
+    if not initial_chunks:
+        raise ValueError("No text chunks extracted from the initial files.")
+
+    initial = initial_chunks[: args.initial_count]
+    if not initial:
+        raise ValueError("Initial chunk set is empty after applying --initial-count.")
+
+    if args.update_files:
+        update_chunks = load_chunks_from_files(list(args.update_files))
+        if not update_chunks:
+            raise ValueError("No text chunks extracted from the update files.")
+        to_add = update_chunks[: args.update_count]
+    else:
+        if not args.update_text:
+            raise ValueError("Provide --update-files or --update-text for the update step.")
+        to_add = [args.update_text]
+    if not to_add:
+        raise ValueError("Update chunk set is empty after applying --update-count.")
+
+    recompute_stats = run_workflow(
+        label="recompute",
+        index_path=args.index_path,
+        initial_paragraphs=initial,
+        update_paragraphs=to_add,
+        model_name=args.embedding_model,
+        embedding_mode=args.embedding_mode,
+        is_recompute=True,
+        query=args.query,
+        top_k=args.top_k,
+    )
+
+    print_results("initial search", recompute_stats["before_results"])
+    print_results("after update", recompute_stats["after_results"])
+    print(
+        f"\n[recompute] Index file size change: {recompute_stats['initial_size']} -> {recompute_stats['updated_size']} bytes"
+        f" (Δ {recompute_stats['delta']})"
+    )
+
+    if recompute_stats["metadata"]:
+        meta_view = {k: recompute_stats["metadata"].get(k) for k in ("is_compact", "is_pruned")}
+        print("[recompute] metadata snapshot:")
+        print(json.dumps(meta_view, indent=2))
+
+    if args.compare_no_recompute:
+        baseline_path = (
+            args.index_path.parent / f"{args.index_path.stem}-norecompute{args.index_path.suffix}"
+        )
+        baseline_stats = run_workflow(
+            label="no-recompute",
+            index_path=baseline_path,
+            initial_paragraphs=initial,
+            update_paragraphs=to_add,
+            model_name=args.embedding_model,
+            embedding_mode=args.embedding_mode,
+            is_recompute=False,
+            query=args.query,
+            top_k=args.top_k,
+        )
+
+        print(
+            f"\n[no-recompute] Index file size change: {baseline_stats['initial_size']} -> {baseline_stats['updated_size']} bytes"
+            f" (Δ {baseline_stats['delta']})"
+        )
+
+        after_texts = [res.text for res in recompute_stats["after_results"]]
+        baseline_after_texts = [res.text for res in baseline_stats["after_results"]]
+        if after_texts == baseline_after_texts:
+            print(
+                "[no-recompute] Search results match recompute baseline; see above for the shared output."
+            )
+        else:
+            print("[no-recompute] WARNING: search results differ from recompute baseline.")
+
+        if baseline_stats["metadata"]:
+            meta_view = {k: baseline_stats["metadata"].get(k) for k in ("is_compact", "is_pruned")}
+            print("[no-recompute] metadata snapshot:")
+            print(json.dumps(meta_view, indent=2))
+
+
+if __name__ == "__main__":
+    main()

packages/leann-backend-diskann/leann_backend_diskann/diskann_embedding_server.py

@@ -10,7 +10,7 @@ import sys
 import threading
 import time
 from pathlib import Path
-from typing import Optional
+from typing import Any, Optional
 
 import numpy as np
 import zmq
@@ -32,6 +32,16 @@ if not logger.handlers:
     logger.propagate = False
 
 
+_RAW_PROVIDER_OPTIONS = os.getenv("LEANN_EMBEDDING_OPTIONS")
+try:
+    PROVIDER_OPTIONS: dict[str, Any] = (
+        json.loads(_RAW_PROVIDER_OPTIONS) if _RAW_PROVIDER_OPTIONS else {}
+    )
+except json.JSONDecodeError:
+    logger.warning("Failed to parse LEANN_EMBEDDING_OPTIONS; ignoring provider options")
+    PROVIDER_OPTIONS = {}
+
+
 def create_diskann_embedding_server(
     passages_file: Optional[str] = None,
     zmq_port: int = 5555,
@@ -181,7 +191,12 @@ def create_diskann_embedding_server(
                     logger.debug(f"Text lengths: {[len(t) for t in texts[:5]]}")  # Show first 5
 
                 # Process embeddings using unified computation
-                embeddings = compute_embeddings(texts, model_name, mode=embedding_mode)
+                embeddings = compute_embeddings(
+                    texts,
+                    model_name,
+                    mode=embedding_mode,
+                    provider_options=PROVIDER_OPTIONS,
+                )
                 logger.info(
                     f"Computed embeddings for {len(texts)} texts, shape: {embeddings.shape}"
                 )
@@ -296,7 +311,12 @@ def create_diskann_embedding_server(
                             continue
 
                     # Process the request
-                    embeddings = compute_embeddings(texts, model_name, mode=embedding_mode)
+                    embeddings = compute_embeddings(
+                        texts,
+                        model_name,
+                        mode=embedding_mode,
+                        provider_options=PROVIDER_OPTIONS,
+                    )
                     logger.info(f"Computed embeddings shape: {embeddings.shape}")
 
                     # Validation

packages/leann-backend-diskann/pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["scikit-build-core>=0.10", "pybind11>=2.12.0", "numpy"]
+requires = ["scikit-build-core>=0.10", "pybind11>=2.12.0", "numpy", "cmake>=3.30"]
 build-backend = "scikit_build_core.build"
 
 [project]

packages/leann-backend-hnsw/leann_backend_hnsw/convert_to_csr.py

@@ -5,6 +5,8 @@ import os
 import struct
 import sys
 import time
+from dataclasses import dataclass
+from typing import Any, Optional
 
 import numpy as np
 
@@ -237,6 +239,288 @@ def write_compact_format(
         f_out.write(storage_data)
 
 
+@dataclass
+class HNSWComponents:
+    original_hnsw_data: dict[str, Any]
+    assign_probas_np: np.ndarray
+    cum_nneighbor_per_level_np: np.ndarray
+    levels_np: np.ndarray
+    is_compact: bool
+    compact_level_ptr: Optional[np.ndarray] = None
+    compact_node_offsets_np: Optional[np.ndarray] = None
+    compact_neighbors_data: Optional[list[int]] = None
+    offsets_np: Optional[np.ndarray] = None
+    neighbors_np: Optional[np.ndarray] = None
+    storage_fourcc: int = NULL_INDEX_FOURCC
+    storage_data: bytes = b""
+
+
+def _read_hnsw_structure(f) -> HNSWComponents:
+    original_hnsw_data: dict[str, Any] = {}
+
+    hnsw_index_fourcc = read_struct(f, "<I")
+    if hnsw_index_fourcc not in EXPECTED_HNSW_FOURCCS:
+        raise ValueError(
+            f"Unexpected HNSW FourCC: {hnsw_index_fourcc:08x}. Expected one of {EXPECTED_HNSW_FOURCCS}."
+        )
+
+    original_hnsw_data["index_fourcc"] = hnsw_index_fourcc
+    original_hnsw_data["d"] = read_struct(f, "<i")
+    original_hnsw_data["ntotal"] = read_struct(f, "<q")
+    original_hnsw_data["dummy1"] = read_struct(f, "<q")
+    original_hnsw_data["dummy2"] = read_struct(f, "<q")
+    original_hnsw_data["is_trained"] = read_struct(f, "?")
+    original_hnsw_data["metric_type"] = read_struct(f, "<i")
+    original_hnsw_data["metric_arg"] = 0.0
+    if original_hnsw_data["metric_type"] > 1:
+        original_hnsw_data["metric_arg"] = read_struct(f, "<f")
+
+    assign_probas_np = read_numpy_vector(f, np.float64, "d")
+    cum_nneighbor_per_level_np = read_numpy_vector(f, np.int32, "i")
+    levels_np = read_numpy_vector(f, np.int32, "i")
+
+    ntotal = len(levels_np)
+    if ntotal != original_hnsw_data["ntotal"]:
+        original_hnsw_data["ntotal"] = ntotal
+
+    pos_before_compact = f.tell()
+    is_compact_flag = None
+    try:
+        is_compact_flag = read_struct(f, "<?")
+    except EOFError:
+        is_compact_flag = None
+
+    if is_compact_flag:
+        compact_level_ptr = read_numpy_vector(f, np.uint64, "Q")
+        compact_node_offsets_np = read_numpy_vector(f, np.uint64, "Q")
+
+        original_hnsw_data["entry_point"] = read_struct(f, "<i")
+        original_hnsw_data["max_level"] = read_struct(f, "<i")
+        original_hnsw_data["efConstruction"] = read_struct(f, "<i")
+        original_hnsw_data["efSearch"] = read_struct(f, "<i")
+        original_hnsw_data["dummy_upper_beam"] = read_struct(f, "<i")
+
+        storage_fourcc = read_struct(f, "<I")
+        compact_neighbors_data_np = read_numpy_vector(f, np.int32, "i")
+        compact_neighbors_data = compact_neighbors_data_np.tolist()
+        storage_data = f.read()
+
+        return HNSWComponents(
+            original_hnsw_data=original_hnsw_data,
+            assign_probas_np=assign_probas_np,
+            cum_nneighbor_per_level_np=cum_nneighbor_per_level_np,
+            levels_np=levels_np,
+            is_compact=True,
+            compact_level_ptr=compact_level_ptr,
+            compact_node_offsets_np=compact_node_offsets_np,
+            compact_neighbors_data=compact_neighbors_data,
+            storage_fourcc=storage_fourcc,
+            storage_data=storage_data,
+        )
+
+    # Non-compact case
+    f.seek(pos_before_compact)
+
+    pos_before_probe = f.tell()
+    try:
+        suspected_flag = read_struct(f, "<B")
+        if suspected_flag != 0x00:
+            f.seek(pos_before_probe)
+    except EOFError:
+        f.seek(pos_before_probe)
+
+    offsets_np = read_numpy_vector(f, np.uint64, "Q")
+    neighbors_np = read_numpy_vector(f, np.int32, "i")
+
+    original_hnsw_data["entry_point"] = read_struct(f, "<i")
+    original_hnsw_data["max_level"] = read_struct(f, "<i")
+    original_hnsw_data["efConstruction"] = read_struct(f, "<i")
+    original_hnsw_data["efSearch"] = read_struct(f, "<i")
+    original_hnsw_data["dummy_upper_beam"] = read_struct(f, "<i")
+
+    storage_fourcc = NULL_INDEX_FOURCC
+    storage_data = b""
+    try:
+        storage_fourcc = read_struct(f, "<I")
+        storage_data = f.read()
+    except EOFError:
+        storage_fourcc = NULL_INDEX_FOURCC
+
+    return HNSWComponents(
+        original_hnsw_data=original_hnsw_data,
+        assign_probas_np=assign_probas_np,
+        cum_nneighbor_per_level_np=cum_nneighbor_per_level_np,
+        levels_np=levels_np,
+        is_compact=False,
+        offsets_np=offsets_np,
+        neighbors_np=neighbors_np,
+        storage_fourcc=storage_fourcc,
+        storage_data=storage_data,
+    )
+
+
+def _read_hnsw_structure_from_file(path: str) -> HNSWComponents:
+    with open(path, "rb") as f:
+        return _read_hnsw_structure(f)
+
+
+def write_original_format(
+    f_out,
+    original_hnsw_data,
+    assign_probas_np,
+    cum_nneighbor_per_level_np,
+    levels_np,
+    offsets_np,
+    neighbors_np,
+    storage_fourcc,
+    storage_data,
+):
+    """Write non-compact HNSW data in original FAISS order."""
+
+    f_out.write(struct.pack("<I", original_hnsw_data["index_fourcc"]))
+    f_out.write(struct.pack("<i", original_hnsw_data["d"]))
+    f_out.write(struct.pack("<q", original_hnsw_data["ntotal"]))
+    f_out.write(struct.pack("<q", original_hnsw_data["dummy1"]))
+    f_out.write(struct.pack("<q", original_hnsw_data["dummy2"]))
+    f_out.write(struct.pack("<?", original_hnsw_data["is_trained"]))
+    f_out.write(struct.pack("<i", original_hnsw_data["metric_type"]))
+    if original_hnsw_data["metric_type"] > 1:
+        f_out.write(struct.pack("<f", original_hnsw_data["metric_arg"]))
+
+    write_numpy_vector(f_out, assign_probas_np, "d")
+    write_numpy_vector(f_out, cum_nneighbor_per_level_np, "i")
+    write_numpy_vector(f_out, levels_np, "i")
+
+    write_numpy_vector(f_out, offsets_np, "Q")
+    write_numpy_vector(f_out, neighbors_np, "i")
+
+    f_out.write(struct.pack("<i", original_hnsw_data["entry_point"]))
+    f_out.write(struct.pack("<i", original_hnsw_data["max_level"]))
+    f_out.write(struct.pack("<i", original_hnsw_data["efConstruction"]))
+    f_out.write(struct.pack("<i", original_hnsw_data["efSearch"]))
+    f_out.write(struct.pack("<i", original_hnsw_data["dummy_upper_beam"]))
+
+    f_out.write(struct.pack("<I", storage_fourcc))
+    if storage_fourcc != NULL_INDEX_FOURCC and storage_data:
+        f_out.write(storage_data)
+
+
+def prune_hnsw_embeddings(input_filename: str, output_filename: str) -> bool:
+    """Rewrite an HNSW index while dropping the embedded storage section."""
+
+    start_time = time.time()
+    try:
+        with open(input_filename, "rb") as f_in, open(output_filename, "wb") as f_out:
+            original_hnsw_data: dict[str, Any] = {}
+
+            hnsw_index_fourcc = read_struct(f_in, "<I")
+            if hnsw_index_fourcc not in EXPECTED_HNSW_FOURCCS:
+                print(
+                    f"Error: Expected HNSW Index FourCC ({list(EXPECTED_HNSW_FOURCCS)}), got {hnsw_index_fourcc:08x}.",
+                    file=sys.stderr,
+                )
+                return False
+
+            original_hnsw_data["index_fourcc"] = hnsw_index_fourcc
+            original_hnsw_data["d"] = read_struct(f_in, "<i")
+            original_hnsw_data["ntotal"] = read_struct(f_in, "<q")
+            original_hnsw_data["dummy1"] = read_struct(f_in, "<q")
+            original_hnsw_data["dummy2"] = read_struct(f_in, "<q")
+            original_hnsw_data["is_trained"] = read_struct(f_in, "?")
+            original_hnsw_data["metric_type"] = read_struct(f_in, "<i")
+            original_hnsw_data["metric_arg"] = 0.0
+            if original_hnsw_data["metric_type"] > 1:
+                original_hnsw_data["metric_arg"] = read_struct(f_in, "<f")
+
+            assign_probas_np = read_numpy_vector(f_in, np.float64, "d")
+            cum_nneighbor_per_level_np = read_numpy_vector(f_in, np.int32, "i")
+            levels_np = read_numpy_vector(f_in, np.int32, "i")
+
+            ntotal = len(levels_np)
+            if ntotal != original_hnsw_data["ntotal"]:
+                original_hnsw_data["ntotal"] = ntotal
+
+            pos_before_compact = f_in.tell()
+            is_compact_flag = None
+            try:
+                is_compact_flag = read_struct(f_in, "<?")
+            except EOFError:
+                is_compact_flag = None
+
+            if is_compact_flag:
+                compact_level_ptr = read_numpy_vector(f_in, np.uint64, "Q")
+                compact_node_offsets_np = read_numpy_vector(f_in, np.uint64, "Q")
+
+                original_hnsw_data["entry_point"] = read_struct(f_in, "<i")
+                original_hnsw_data["max_level"] = read_struct(f_in, "<i")
+                original_hnsw_data["efConstruction"] = read_struct(f_in, "<i")
+                original_hnsw_data["efSearch"] = read_struct(f_in, "<i")
+                original_hnsw_data["dummy_upper_beam"] = read_struct(f_in, "<i")
+
+                _storage_fourcc = read_struct(f_in, "<I")
+                compact_neighbors_data_np = read_numpy_vector(f_in, np.int32, "i")
+                compact_neighbors_data = compact_neighbors_data_np.tolist()
+                _storage_data = f_in.read()
+
+                write_compact_format(
+                    f_out,
+                    original_hnsw_data,
+                    assign_probas_np,
+                    cum_nneighbor_per_level_np,
+                    levels_np,
+                    compact_level_ptr,
+                    compact_node_offsets_np,
+                    compact_neighbors_data,
+                    NULL_INDEX_FOURCC,
+                    b"",
+                )
+            else:
+                f_in.seek(pos_before_compact)
+
+                pos_before_probe = f_in.tell()
+                try:
+                    suspected_flag = read_struct(f_in, "<B")
+                    if suspected_flag != 0x00:
+                        f_in.seek(pos_before_probe)
+                except EOFError:
+                    f_in.seek(pos_before_probe)
+
+                offsets_np = read_numpy_vector(f_in, np.uint64, "Q")
+                neighbors_np = read_numpy_vector(f_in, np.int32, "i")
+
+                original_hnsw_data["entry_point"] = read_struct(f_in, "<i")
+                original_hnsw_data["max_level"] = read_struct(f_in, "<i")
+                original_hnsw_data["efConstruction"] = read_struct(f_in, "<i")
+                original_hnsw_data["efSearch"] = read_struct(f_in, "<i")
+                original_hnsw_data["dummy_upper_beam"] = read_struct(f_in, "<i")
+
+                _storage_fourcc = None
+                _storage_data = b""
+                try:
+                    _storage_fourcc = read_struct(f_in, "<I")
+                    _storage_data = f_in.read()
+                except EOFError:
+                    _storage_fourcc = NULL_INDEX_FOURCC
+
+                write_original_format(
+                    f_out,
+                    original_hnsw_data,
+                    assign_probas_np,
+                    cum_nneighbor_per_level_np,
+                    levels_np,
+                    offsets_np,
+                    neighbors_np,
+                    NULL_INDEX_FOURCC,
+                    b"",
+                )
+
+        print(f"[{time.time() - start_time:.2f}s] Pruned embeddings from {input_filename}")
+        return True
+    except Exception as exc:
+        print(f"Failed to prune embeddings: {exc}", file=sys.stderr)
+        return False
+
+
 # --- Main Conversion Logic ---
 
 
@@ -700,6 +984,29 @@ def convert_hnsw_graph_to_csr(input_filename, output_filename, prune_embeddings=
             pass
 
 
+def prune_hnsw_embeddings_inplace(index_filename: str) -> bool:
+    """Convenience wrapper to prune embeddings in-place."""
+
+    temp_path = f"{index_filename}.prune.tmp"
+    success = prune_hnsw_embeddings(index_filename, temp_path)
+    if success:
+        try:
+            os.replace(temp_path, index_filename)
+        except Exception as exc:  # pragma: no cover - defensive
+            logger.error(f"Failed to replace original index with pruned version: {exc}")
+            try:
+                os.remove(temp_path)
+            except OSError:
+                pass
+            return False
+    else:
+        try:
+            os.remove(temp_path)
+        except OSError:
+            pass
+    return success
+
+
 # --- Script Execution ---
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(

packages/leann-backend-hnsw/leann_backend_hnsw/hnsw_backend.py

@@ -14,8 +14,7 @@ from leann.interface import (
 from leann.registry import register_backend
 from leann.searcher_base import BaseSearcher
 
-from .convert_to_csr import convert_hnsw_graph_to_csr
-from .prune_index import prune_embeddings_preserve_graph_inplace
+from .convert_to_csr import convert_hnsw_graph_to_csr, prune_hnsw_embeddings_inplace
 
 logger = logging.getLogger(__name__)
 
@@ -91,16 +90,10 @@ class HNSWBuilder(LeannBackendBuilderInterface):
         index_file = index_dir / f"{index_prefix}.index"
         faiss.write_index(index, str(index_file))
 
-        if self.is_recompute:
-            if self.is_compact:
-                self._convert_to_csr(index_file)
-            else:
-                # Non-compact format: prune only embeddings, keep original graph
-                ok = prune_embeddings_preserve_graph_inplace(str(index_file))
-                if not ok:
-                    raise RuntimeError(
-                        "Pruning embeddings while preserving graph failed for non-compact index"
-                    )
+        if self.is_compact:
+            self._convert_to_csr(index_file)
+        elif self.is_recompute:
+            prune_hnsw_embeddings_inplace(str(index_file))
 
     def _convert_to_csr(self, index_file: Path):
         """Convert built index to CSR format"""
@@ -142,10 +135,10 @@ class HNSWSearcher(BaseSearcher):
         if metric_enum is None:
             raise ValueError(f"Unsupported distance_metric '{self.distance_metric}'.")
 
-        self.is_compact, self.is_pruned = (
-            self.meta.get("is_compact", True),
-            self.meta.get("is_pruned", True),
-        )
+        backend_meta_kwargs = self.meta.get("backend_kwargs", {})
+        self.is_compact = self.meta.get("is_compact", backend_meta_kwargs.get("is_compact", True))
+        default_pruned = backend_meta_kwargs.get("is_recompute", self.is_compact)
+        self.is_pruned = bool(self.meta.get("is_pruned", default_pruned))
 
         index_file = self.index_dir / f"{self.index_path.stem}.index"
         if not index_file.exists():
@@ -157,13 +150,7 @@ class HNSWSearcher(BaseSearcher):
             self.is_pruned
         )  # In C++ code, it's called is_recompute, but it's only for loading IIUC.
 
-        # If pruned (recompute mode), explicitly skip storage to avoid reading
-        # the pruned section. Still allow MMAP for graph.
-        io_flags = faiss.IO_FLAG_MMAP
-        if self.is_pruned:
-            io_flags |= faiss.IO_FLAG_SKIP_STORAGE
-
-        self._index = faiss.read_index(str(index_file), io_flags, hnsw_config)
+        self._index = faiss.read_index(str(index_file), faiss.IO_FLAG_MMAP, hnsw_config)
 
     def search(
         self,
@@ -266,55 +253,3 @@ class HNSWSearcher(BaseSearcher):
         string_labels = [[str(int_label) for int_label in batch_labels] for batch_labels in labels]
 
         return {"labels": string_labels, "distances": distances}
-
-
-# ---------- Helper API for incremental add (Python-level) ----------
-def add_vectors(
-    index_file_path: str,
-    embeddings: np.ndarray,
-    *,
-    ef_construction: Optional[int] = None,
-    recompute: bool = False,
-) -> None:
-    """Append vectors to an existing non-compact HNSW index.
-
-    Args:
-        index_file_path: Path to the HNSW .index file
-        embeddings: float32 numpy array (N, D)
-        ef_construction: Optional override for efConstruction during insertion
-        recompute: Reserved for future use to control insertion-time recompute behaviors
-    """
-    from . import faiss  # type: ignore
-
-    if embeddings.dtype != np.float32:
-        embeddings = embeddings.astype(np.float32)
-    if not embeddings.flags.c_contiguous:
-        embeddings = np.ascontiguousarray(embeddings, dtype=np.float32)
-
-    # Load index normally to ensure storage is present; toggle is_recompute on the object
-    index = faiss.read_index(str(index_file_path), faiss.IO_FLAG_MMAP)
-
-    # Best-effort: explicitly set flag on the object if the binding exposes it
-    try:
-        index.is_recompute = bool(recompute)
-    except Exception:
-        pass
-    try:
-        if ef_construction is not None:
-            index.hnsw.efConstruction = int(ef_construction)
-    except Exception:
-        # Best-effort; ignore if backend doesn't expose setter
-        pass
-
-    # For non-compact HNSW, calling add directly is sufficient. When is_recompute is set
-    # (via config or attribute), FAISS will run the insertion/search path accordingly.
-    # To strictly follow per-point insert semantics in recompute mode, add one-by-one.
-    if recompute:
-        # Insert row by row
-        n = embeddings.shape[0]
-        for i in range(n):
-            row = embeddings[i : i + 1]
-            index.add(1, faiss.swig_ptr(row))
-    else:
-        index.add(embeddings.shape[0], faiss.swig_ptr(embeddings))
-    faiss.write_index(index, str(index_file_path))

packages/leann-backend-hnsw/leann_backend_hnsw/hnsw_embedding_server.py

@@ -10,7 +10,7 @@ import sys
 import threading
 import time
 from pathlib import Path
-from typing import Optional
+from typing import Any, Optional
 
 import msgpack
 import numpy as np
@@ -24,13 +24,35 @@ logger = logging.getLogger(__name__)
 log_level = getattr(logging, LOG_LEVEL, logging.WARNING)
 logger.setLevel(log_level)
 
-# Ensure we have a handler if none exists
+# Ensure we have handlers if none exist
 if not logger.handlers:
-    handler = logging.StreamHandler()
+    stream_handler = logging.StreamHandler()
     formatter = logging.Formatter("%(asctime)s - %(levelname)s - %(message)s")
-    handler.setFormatter(formatter)
-    logger.addHandler(handler)
-    logger.propagate = False
+    stream_handler.setFormatter(formatter)
+    logger.addHandler(stream_handler)
+
+log_path = os.getenv("LEANN_HNSW_LOG_PATH")
+if log_path:
+    try:
+        file_handler = logging.FileHandler(log_path, mode="a", encoding="utf-8")
+        file_formatter = logging.Formatter(
+            "%(asctime)s - %(levelname)s - [pid=%(process)d] %(message)s"
+        )
+        file_handler.setFormatter(file_formatter)
+        logger.addHandler(file_handler)
+    except Exception as exc:  # pragma: no cover - best effort logging
+        logger.warning(f"Failed to attach file handler for log path {log_path}: {exc}")
+
+logger.propagate = False
+
+_RAW_PROVIDER_OPTIONS = os.getenv("LEANN_EMBEDDING_OPTIONS")
+try:
+    PROVIDER_OPTIONS: dict[str, Any] = (
+        json.loads(_RAW_PROVIDER_OPTIONS) if _RAW_PROVIDER_OPTIONS else {}
+    )
+except json.JSONDecodeError:
+    logger.warning("Failed to parse LEANN_EMBEDDING_OPTIONS; ignoring provider options")
+    PROVIDER_OPTIONS = {}
 
 
 def create_hnsw_embedding_server(
@@ -138,7 +160,12 @@ def create_hnsw_embedding_server(
                     ):
                         last_request_type = "text"
                         last_request_length = len(request)
-                        embeddings = compute_embeddings(request, model_name, mode=embedding_mode)
+                        embeddings = compute_embeddings(
+                            request,
+                            model_name,
+                            mode=embedding_mode,
+                            provider_options=PROVIDER_OPTIONS,
+                        )
                         rep_socket.send(msgpack.packb(embeddings.tolist()))
                         e2e_end = time.time()
                         logger.info(f"⏱️  Text embedding E2E time: {e2e_end - e2e_start:.6f}s")
@@ -187,7 +214,10 @@ def create_hnsw_embedding_server(
                         if texts:
                             try:
                                 embeddings = compute_embeddings(
-                                    texts, model_name, mode=embedding_mode
+                                    texts,
+                                    model_name,
+                                    mode=embedding_mode,
+                                    provider_options=PROVIDER_OPTIONS,
                                 )
                                 logger.info(
                                     f"Computed embeddings for {len(texts)} texts, shape: {embeddings.shape}"
@@ -252,7 +282,12 @@ def create_hnsw_embedding_server(
 
                     if texts:
                         try:
-                            embeddings = compute_embeddings(texts, model_name, mode=embedding_mode)
+                            embeddings = compute_embeddings(
+                                texts,
+                                model_name,
+                                mode=embedding_mode,
+                                provider_options=PROVIDER_OPTIONS,
+                            )
                             logger.info(
                                 f"Computed embeddings for {len(texts)} texts, shape: {embeddings.shape}"
                             )

packages/leann-backend-hnsw/leann_backend_hnsw/prune_index.py

@@ -1,149 +0,0 @@
-import os
-import struct
-from pathlib import Path
-
-from .convert_to_csr import (
-    EXPECTED_HNSW_FOURCCS,
-    NULL_INDEX_FOURCC,
-    read_struct,
-    read_vector_raw,
-)
-
-
-def _write_vector_raw(f_out, count: int, data_bytes: bytes) -> None:
-    """Write a vector in the same binary layout as read_vector_raw reads: <Q count> + raw bytes."""
-    f_out.write(struct.pack("<Q", count))
-    if count > 0 and data_bytes:
-        f_out.write(data_bytes)
-
-
-def prune_embeddings_preserve_graph(input_filename: str, output_filename: str) -> bool:
-    """
-    Copy an original (non-compact) HNSW index file while pruning the trailing embedding storage.
-    Preserves the graph structure and metadata exactly; only writes a NULL storage marker instead of
-    the original storage fourcc and payload.
-
-    Returns True on success.
-    """
-    print(f"Pruning embeddings from {input_filename} to {output_filename}")
-    print("--------------------------------")
-    # running in mode is-recompute=True and is-compact=False
-    in_path = Path(input_filename)
-    out_path = Path(output_filename)
-
-    try:
-        with open(in_path, "rb") as f_in, open(out_path, "wb") as f_out:
-            # Header
-            index_fourcc = read_struct(f_in, "<I")
-            if index_fourcc not in EXPECTED_HNSW_FOURCCS:
-                # Still proceed, but this is unexpected
-                pass
-            f_out.write(struct.pack("<I", index_fourcc))
-
-            d = read_struct(f_in, "<i")
-            ntotal_hdr = read_struct(f_in, "<q")
-            dummy1 = read_struct(f_in, "<q")
-            dummy2 = read_struct(f_in, "<q")
-            is_trained = read_struct(f_in, "?")
-            metric_type = read_struct(f_in, "<i")
-            f_out.write(struct.pack("<i", d))
-            f_out.write(struct.pack("<q", ntotal_hdr))
-            f_out.write(struct.pack("<q", dummy1))
-            f_out.write(struct.pack("<q", dummy2))
-            f_out.write(struct.pack("<?", is_trained))
-            f_out.write(struct.pack("<i", metric_type))
-
-            if metric_type > 1:
-                metric_arg = read_struct(f_in, "<f")
-                f_out.write(struct.pack("<f", metric_arg))
-
-            # Vectors: assign_probas (double), cum_nneighbor_per_level (int32), levels (int32)
-            cnt, data = read_vector_raw(f_in, "d")
-            _write_vector_raw(f_out, cnt, data)
-
-            cnt, data = read_vector_raw(f_in, "i")
-            _write_vector_raw(f_out, cnt, data)
-
-            cnt, data = read_vector_raw(f_in, "i")
-            _write_vector_raw(f_out, cnt, data)
-
-            # Probe potential extra alignment/flag byte present in some original formats
-            probe = f_in.read(1)
-            if probe:
-                if probe == b"\x00":
-                    # Preserve this unexpected 0x00 byte
-                    f_out.write(probe)
-                else:
-                    # Likely part of the next vector; rewind
-                    f_in.seek(-1, os.SEEK_CUR)
-
-            # Offsets (uint64) and neighbors (int32)
-            cnt, data = read_vector_raw(f_in, "Q")
-            _write_vector_raw(f_out, cnt, data)
-
-            cnt, data = read_vector_raw(f_in, "i")
-            _write_vector_raw(f_out, cnt, data)
-
-            # Scalar params
-            entry_point = read_struct(f_in, "<i")
-            max_level = read_struct(f_in, "<i")
-            ef_construction = read_struct(f_in, "<i")
-            ef_search = read_struct(f_in, "<i")
-            dummy_upper_beam = read_struct(f_in, "<i")
-            f_out.write(struct.pack("<i", entry_point))
-            f_out.write(struct.pack("<i", max_level))
-            f_out.write(struct.pack("<i", ef_construction))
-            f_out.write(struct.pack("<i", ef_search))
-            f_out.write(struct.pack("<i", dummy_upper_beam))
-
-            # Storage fourcc (if present) — write NULL marker and drop any remaining data
-            try:
-                read_struct(f_in, "<I")
-                # Regardless of original, write NULL
-                f_out.write(struct.pack("<I", NULL_INDEX_FOURCC))
-                # Discard the rest of the file (embedding payload)
-                # (Do not copy anything else)
-            except EOFError:
-                # No storage section; nothing else to write
-                pass
-
-        return True
-    except Exception:
-        # Best-effort cleanup
-        try:
-            if out_path.exists():
-                out_path.unlink()
-        except OSError:
-            pass
-        return False
-
-
-def prune_embeddings_preserve_graph_inplace(index_file_path: str) -> bool:
-    """
-    Convenience wrapper: write pruned file to a temporary path next to the
-    original, then atomically replace on success.
-    """
-    print(f"Pruning embeddings from {index_file_path} to {index_file_path}")
-    print("--------------------------------")
-    # running in mode is-recompute=True and is-compact=False
-    src = Path(index_file_path)
-    tmp = src.with_suffix(".pruned.tmp")
-    ok = prune_embeddings_preserve_graph(str(src), str(tmp))
-    if not ok:
-        if tmp.exists():
-            try:
-                tmp.unlink()
-            except OSError:
-                pass
-        return False
-    try:
-        os.replace(str(tmp), str(src))
-    except Exception:
-        # Rollback on failure
-        try:
-            if tmp.exists():
-                tmp.unlink()
-        except OSError:
-            pass
-        return False
-    return True

packages/leann-core/src/leann/api.py

@@ -5,7 +5,6 @@ with the correct, original embedding logic from the user's reference code.
 
 import json
 import logging
-import os
 import pickle
 import re
 import subprocess
@@ -16,11 +15,11 @@ from pathlib import Path
 from typing import Any, Literal, Optional, Union
 
 import numpy as np
+from leann_backend_hnsw.convert_to_csr import prune_hnsw_embeddings_inplace
 
 from leann.interface import LeannBackendSearcherInterface
 
 from .chat import get_llm
-from .embedding_server_manager import EmbeddingServerManager
 from .interface import LeannBackendFactoryInterface
 from .metadata_filter import MetadataFilterEngine
 from .registry import BACKEND_REGISTRY
@@ -40,6 +39,7 @@ def compute_embeddings(
     use_server: bool = True,
     port: Optional[int] = None,
     is_build=False,
+    provider_options: Optional[dict[str, Any]] = None,
 ) -> np.ndarray:
     """
     Computes embeddings using different backends.
@@ -73,6 +73,7 @@ def compute_embeddings(
             model_name,
             mode=mode,
             is_build=is_build,
+            provider_options=provider_options,
         )
 
 
@@ -120,20 +121,6 @@ class SearchResult:
     metadata: dict[str, Any] = field(default_factory=dict)
 
 
-@dataclass
-class IncrementalAddContext:
-    """Prepared context for safe incremental add operations on an index."""
-
-    index_path: str
-    passages_file: Path
-    offsets_file: Path
-    vector_index_file: Path
-    embedding_model: str
-    embedding_mode: str
-    distance_metric: str
-    prepared_texts: Optional[list[str]] = None
-
-
 class PassageManager:
     def __init__(
         self, passage_sources: list[dict[str, Any]], metadata_file_path: Optional[str] = None
@@ -293,6 +280,7 @@ class LeannBuilder:
         embedding_model: str = "facebook/contriever",
         dimensions: Optional[int] = None,
         embedding_mode: str = "sentence-transformers",
+        embedding_options: Optional[dict[str, Any]] = None,
         **backend_kwargs,
     ):
         self.backend_name = backend_name
@@ -315,6 +303,7 @@ class LeannBuilder:
         self.embedding_model = embedding_model
         self.dimensions = dimensions
         self.embedding_mode = embedding_mode
+        self.embedding_options = embedding_options or {}
 
         # Check if we need to use cosine distance for normalized embeddings
         normalized_embeddings_models = {
@@ -422,6 +411,7 @@ class LeannBuilder:
                     self.embedding_model,
                     self.embedding_mode,
                     use_server=False,
+                    provider_options=self.embedding_options,
                 )[0]
             )
         path = Path(index_path)
@@ -461,6 +451,7 @@ class LeannBuilder:
             self.embedding_mode,
             use_server=False,
             is_build=True,
+            provider_options=self.embedding_options,
         )
         string_ids = [chunk["id"] for chunk in self.chunks]
         current_backend_kwargs = {**self.backend_kwargs, "dimensions": self.dimensions}
@@ -487,12 +478,15 @@ class LeannBuilder:
             ],
         }
 
+        if self.embedding_options:
+            meta_data["embedding_options"] = self.embedding_options
+
         # Add storage status flags for HNSW backend
         if self.backend_name == "hnsw":
             is_compact = self.backend_kwargs.get("is_compact", True)
             is_recompute = self.backend_kwargs.get("is_recompute", True)
             meta_data["is_compact"] = is_compact
-            meta_data["is_pruned"] = is_recompute  # Pruned only if compact and recompute
+            meta_data["is_pruned"] = bool(is_recompute)
         with open(leann_meta_path, "w", encoding="utf-8") as f:
             json.dump(meta_data, f, indent=2)
 
@@ -607,18 +601,166 @@ class LeannBuilder:
             "embeddings_source": str(embeddings_file),
         }
 
+        if self.embedding_options:
+            meta_data["embedding_options"] = self.embedding_options
+
         # Add storage status flags for HNSW backend
         if self.backend_name == "hnsw":
             is_compact = self.backend_kwargs.get("is_compact", True)
             is_recompute = self.backend_kwargs.get("is_recompute", True)
             meta_data["is_compact"] = is_compact
-            meta_data["is_pruned"] = is_compact and is_recompute
+            meta_data["is_pruned"] = bool(is_recompute)
 
         with open(leann_meta_path, "w", encoding="utf-8") as f:
             json.dump(meta_data, f, indent=2)
 
         logger.info(f"Index built successfully from precomputed embeddings: {index_path}")
 
+    def update_index(self, index_path: str):
+        """Append new passages and vectors to an existing HNSW index."""
+        if not self.chunks:
+            raise ValueError("No new chunks provided for update.")
+
+        path = Path(index_path)
+        index_dir = path.parent
+        index_name = path.name
+        index_prefix = path.stem
+
+        meta_path = index_dir / f"{index_name}.meta.json"
+        passages_file = index_dir / f"{index_name}.passages.jsonl"
+        offset_file = index_dir / f"{index_name}.passages.idx"
+        index_file = index_dir / f"{index_prefix}.index"
+
+        if not meta_path.exists() or not passages_file.exists() or not offset_file.exists():
+            raise FileNotFoundError("Index metadata or passage files are missing; cannot update.")
+        if not index_file.exists():
+            raise FileNotFoundError(f"HNSW index file not found: {index_file}")
+
+        with open(meta_path, encoding="utf-8") as f:
+            meta = json.load(f)
+        backend_name = meta.get("backend_name")
+        if backend_name != self.backend_name:
+            raise ValueError(
+                f"Index was built with backend '{backend_name}', cannot update with '{self.backend_name}'."
+            )
+
+        meta_backend_kwargs = meta.get("backend_kwargs", {})
+        index_is_compact = meta.get("is_compact", meta_backend_kwargs.get("is_compact", True))
+        if index_is_compact:
+            raise ValueError(
+                "Compact HNSW indices do not support in-place updates. Rebuild required."
+            )
+
+        distance_metric = meta_backend_kwargs.get(
+            "distance_metric", self.backend_kwargs.get("distance_metric", "mips")
+        ).lower()
+        needs_recompute = bool(
+            meta.get("is_pruned")
+            or meta_backend_kwargs.get("is_recompute")
+            or self.backend_kwargs.get("is_recompute")
+        )
+
+        with open(offset_file, "rb") as f:
+            offset_map: dict[str, int] = pickle.load(f)
+        existing_ids = set(offset_map.keys())
+
+        valid_chunks: list[dict[str, Any]] = []
+        for chunk in self.chunks:
+            text = chunk.get("text", "")
+            if not isinstance(text, str) or not text.strip():
+                continue
+            metadata = chunk.setdefault("metadata", {})
+            passage_id = chunk.get("id") or metadata.get("id")
+            if passage_id and passage_id in existing_ids:
+                raise ValueError(f"Passage ID '{passage_id}' already exists in the index.")
+            valid_chunks.append(chunk)
+
+        if not valid_chunks:
+            raise ValueError("No valid chunks to append.")
+
+        texts_to_embed = [chunk["text"] for chunk in valid_chunks]
+        embeddings = compute_embeddings(
+            texts_to_embed,
+            self.embedding_model,
+            self.embedding_mode,
+            use_server=False,
+            is_build=True,
+            provider_options=self.embedding_options,
+        )
+
+        embedding_dim = embeddings.shape[1]
+        expected_dim = meta.get("dimensions")
+        if expected_dim is not None and expected_dim != embedding_dim:
+            raise ValueError(
+                f"Dimension mismatch during update: existing index uses {expected_dim}, got {embedding_dim}."
+            )
+
+        from leann_backend_hnsw import faiss  # type: ignore
+
+        embeddings = np.ascontiguousarray(embeddings, dtype=np.float32)
+        if distance_metric == "cosine":
+            norms = np.linalg.norm(embeddings, axis=1, keepdims=True)
+            norms[norms == 0] = 1
+            embeddings = embeddings / norms
+
+        index = faiss.read_index(str(index_file))
+        if hasattr(index, "is_recompute"):
+            index.is_recompute = needs_recompute
+        if getattr(index, "storage", None) is None:
+            if index.metric_type == faiss.METRIC_INNER_PRODUCT:
+                storage_index = faiss.IndexFlatIP(index.d)
+            else:
+                storage_index = faiss.IndexFlatL2(index.d)
+            index.storage = storage_index
+            index.own_fields = True
+        if index.d != embedding_dim:
+            raise ValueError(
+                f"Existing index dimension ({index.d}) does not match new embeddings ({embedding_dim})."
+            )
+
+        base_id = index.ntotal
+        for offset, chunk in enumerate(valid_chunks):
+            new_id = str(base_id + offset)
+            chunk.setdefault("metadata", {})["id"] = new_id
+            chunk["id"] = new_id
+
+        index.add(embeddings.shape[0], faiss.swig_ptr(embeddings))
+        faiss.write_index(index, str(index_file))
+
+        with open(passages_file, "a", encoding="utf-8") as f:
+            for chunk in valid_chunks:
+                offset = f.tell()
+                json.dump(
+                    {
+                        "id": chunk["id"],
+                        "text": chunk["text"],
+                        "metadata": chunk.get("metadata", {}),
+                    },
+                    f,
+                    ensure_ascii=False,
+                )
+                f.write("\n")
+                offset_map[chunk["id"]] = offset
+
+        with open(offset_file, "wb") as f:
+            pickle.dump(offset_map, f)
+
+        meta["total_passages"] = len(offset_map)
+        with open(meta_path, "w", encoding="utf-8") as f:
+            json.dump(meta, f, indent=2)
+
+        logger.info(
+            "Appended %d passages to index '%s'. New total: %d",
+            len(valid_chunks),
+            index_path,
+            len(offset_map),
+        )
+
+        self.chunks.clear()
+
+        if needs_recompute:
+            prune_hnsw_embeddings_inplace(str(index_file))
+
 
 class LeannSearcher:
     def __init__(self, index_path: str, enable_warmup: bool = False, **backend_kwargs):
@@ -642,6 +784,7 @@ class LeannSearcher:
         self.embedding_model = self.meta_data["embedding_model"]
         # Support both old and new format
         self.embedding_mode = self.meta_data.get("embedding_mode", "sentence-transformers")
+        self.embedding_options = self.meta_data.get("embedding_options", {})
         # Delegate portability handling to PassageManager
         self.passage_manager = PassageManager(
             self.meta_data.get("passage_sources", []), metadata_file_path=self.meta_path_str
@@ -653,6 +796,8 @@ class LeannSearcher:
             raise ValueError(f"Backend '{backend_name}' not found.")
         final_kwargs = {**self.meta_data.get("backend_kwargs", {}), **backend_kwargs}
         final_kwargs["enable_warmup"] = enable_warmup
+        if self.embedding_options:
+            final_kwargs.setdefault("embedding_options", self.embedding_options)
         self.backend_impl: LeannBackendSearcherInterface = backend_factory.searcher(
             index_path, **final_kwargs
         )
@@ -1032,405 +1177,8 @@ class LeannChat:
         except Exception:
             pass
 
-
-# ------------------------------
-# Incremental Add Utilities (HNSW no-recompute only)
-# ------------------------------
-
-
-def _resolve_index_paths(index_path: str) -> tuple[Path, Path, Path]:
-    """Given base index path (without extension), return (passages.jsonl, passages.idx, vector.index).
-
-    For HNSW, vector index file is typically <stem>.index (e.g., documents.index) even when base is
-    'documents.leann'. We prefer an existing <stem>.index, otherwise fall back to <name>.index.
-    """
-    base = Path(index_path)
-    passages_file = base.parent / f"{base.name}.passages.jsonl"
-    offsets_file = base.parent / f"{base.name}.passages.idx"
-    candidate_name_index = base.parent / f"{base.name}.index"
-    candidate_stem_index = base.parent / f"{base.stem}.index"
-    vector_index_file = (
-        candidate_stem_index if candidate_stem_index.exists() else candidate_name_index
-    )
-    return passages_file, offsets_file, vector_index_file
-
-
-def _read_meta_file(index_path: str) -> dict[str, Any]:
-    meta_path = Path(f"{index_path}.meta.json")
-    if not meta_path.exists():
-        raise FileNotFoundError(f"Leann metadata file not found: {meta_path}")
-    with open(meta_path, encoding="utf-8") as f:
-        return json.load(f)
-
-
-def _load_offset_map_pickle(offsets_file: Path) -> dict[str, int]:
-    if not offsets_file.exists():
-        return {}
-    with open(offsets_file, "rb") as f:
-        return pickle.load(f)
-
-
-def _append_passages_and_update_offsets(
-    passages_file: Path, offsets_file: Path, new_texts: list[str]
-) -> list[str]:
-    """Append new texts to passages file, update offset map, and return assigned string IDs.
-
-    IDs are assigned as incrementing integers based on existing keys in the offset map.
-    """
-    offset_map = _load_offset_map_pickle(offsets_file)
-    # Compute next numeric id
-    numeric_ids = [int(x) for x in offset_map.keys() if str(x).isdigit()]
-    next_id_num = (max(numeric_ids) + 1) if numeric_ids else 0
-    assigned_ids: list[str] = []
-
-    with open(passages_file, "a", encoding="utf-8") as f:
-        for text in new_texts:
-            offset = f.tell()
-            str_id = str(next_id_num)
-            json.dump({"id": str_id, "text": text, "metadata": {}}, f, ensure_ascii=False)
-            f.write("\n")
-            offset_map[str_id] = offset
-            assigned_ids.append(str_id)
-            next_id_num += 1
-
-    with open(offsets_file, "wb") as f:
-        pickle.dump(offset_map, f)
-
-    return assigned_ids
-
-
-def incremental_add_texts(
-    index_path: str,
-    texts: list[str],
-    *,
-    embedding_model: Optional[str] = None,
-    embedding_mode: Optional[str] = None,
-    ef_construction: Optional[int] = None,
-    recompute: bool = False,
-) -> int:
-    """Incrementally add text chunks to an existing HNSW index built with no-recompute.
-
-    - Validates backend is HNSW and index is non-compact (no-recompute path)
-    - Appends passages and offsets
-    - Computes embeddings and appends to the HNSW vector index
-
-    Returns number of added chunks.
-    """
-    if not texts:
-        return 0
-
-    meta = _read_meta_file(index_path)
-    if meta.get("backend_name") != "hnsw":
-        raise RuntimeError("Incremental add is currently supported only for HNSW backend")
-    if meta.get("is_compact", True):
-        raise RuntimeError(
-            "Index is compact/pruned. Rebuild base with is_recompute=False and is_compact=False for incremental add."
-        )
-
-    passages_file, offsets_file, vector_index_file = _resolve_index_paths(index_path)
-    if not vector_index_file.exists():
-        raise FileNotFoundError(
-            f"Vector index file missing: {vector_index_file}. Build base first with LeannBuilder."
-        )
-
-    # Resolve embedding config from meta if not provided
-    model_name = embedding_model or meta.get("embedding_model", "facebook/contriever")
-    mode_name = embedding_mode or meta.get("embedding_mode", "sentence-transformers")
-
-    # Append passages and update offsets
-    assigned_ids = _append_passages_and_update_offsets(passages_file, offsets_file, texts)
-
-    # Compute embeddings
-    # Embedding computation path
-    esm = None
-    port = None
-    if recompute:
-        # Determine distance metric early for server config
-        distance_metric = meta.get("backend_kwargs", {}).get("distance_metric", "mips").lower()
-        # Start embedding server and compute via ZMQ for consistency with recompute semantics
-        passages_source_file = f"{index_path}.meta.json"
-        esm = EmbeddingServerManager(
-            backend_module_name="leann_backend_hnsw.hnsw_embedding_server",
-        )
-        started, port = esm.start_server(
-            port=5557,
-            model_name=model_name,
-            embedding_mode=mode_name,
-            passages_file=passages_source_file,
-            distance_metric=distance_metric,
-            enable_warmup=False,
-        )
-        if not started:
-            raise RuntimeError("Failed to start embedding server for recompute add")
-        embeddings = compute_embeddings_via_server(texts, model_name, port)
-    else:
-        embeddings = compute_embeddings(
-            texts,
-            model_name=model_name,
-            mode=mode_name,
-            use_server=False,
-            is_build=True,
-        )
-
-    # Normalize for cosine if needed
-    if "distance_metric" not in locals():
-        distance_metric = meta.get("backend_kwargs", {}).get("distance_metric", "mips").lower()
-    if distance_metric == "cosine":
-        norms = np.linalg.norm(embeddings, axis=1, keepdims=True)
-        norms[norms == 0] = 1
-        embeddings = embeddings / norms
-
-    # Append via backend helper (supports ef_construction/recompute plumbing)
-    try:
-        from leann_backend_hnsw.hnsw_backend import add_vectors as hnsw_add_vectors  # type: ignore
-    except Exception as e:
-        raise RuntimeError(
-            "Failed to import HNSW backend add helper. Ensure HNSW backend is installed."
-        ) from e
-
-    # Propagate ZMQ port to FAISS add path when recompute is True
-    if recompute and port is not None:
-        os.environ["LEANN_ZMQ_PORT"] = str(port)
-
-    hnsw_add_vectors(
-        str(vector_index_file),
-        embeddings,
-        ef_construction=ef_construction,
-        recompute=recompute,
-    )
-
-    # Stop server after add when recompute path used
-    if esm is not None:
+    def __del__(self):
         try:
-            esm.stop_server()
+            self.cleanup()
         except Exception:
             pass
-
-    # Sanity: ids length should match embeddings rows
-    if len(assigned_ids) != embeddings.shape[0]:
-        warnings.warn(
-            f"Assigned {len(assigned_ids)} IDs but computed {embeddings.shape[0]} embeddings.",
-            UserWarning,
-            stacklevel=2,
-        )
-
-    return len(assigned_ids)
-
-
-def create_incremental_add_context(
-    index_path: str,
-    *,
-    # Optional embedding choices; if None will use meta
-    embedding_model: Optional[str] = None,
-    embedding_mode: Optional[str] = None,
-    # Optional data-to-text preparation in context
-    data_dir: Optional[str] = None,
-    required_exts: Optional[list[str]] = None,
-    chunk_size: int = 256,
-    chunk_overlap: int = 128,
-    max_items: int = -1,
-) -> IncrementalAddContext:
-    """Validate index and prepare context for repeated incremental adds.
-
-    Additionally, if data_dir is provided, this function will load documents,
-    chunk them to texts with the specified parameters, and store them in ctx.prepared_texts.
-    """
-    meta = _read_meta_file(index_path)
-    if meta.get("backend_name") != "hnsw":
-        raise RuntimeError("Incremental add is currently supported only for HNSW backend")
-    if meta.get("is_compact", True):
-        raise RuntimeError(
-            "Index is compact/pruned. Rebuild base with is_recompute=False and is_compact=False for incremental add."
-        )
-
-    passages_file, offsets_file, vector_index_file = _resolve_index_paths(index_path)
-    if not vector_index_file.exists():
-        raise FileNotFoundError(
-            f"Vector index file missing: {vector_index_file}. Build base first with LeannBuilder."
-        )
-
-    model_name = embedding_model or meta.get("embedding_model", "facebook/contriever")
-    mode_name = embedding_mode or meta.get("embedding_mode", "sentence-transformers")
-    distance_metric = meta.get("backend_kwargs", {}).get("distance_metric", "mips").lower()
-
-    prepared_texts: Optional[list[str]] = None
-    if data_dir is not None:
-        try:
-            from llama_index.core import SimpleDirectoryReader  # type: ignore
-            from llama_index.core.node_parser import SentenceSplitter  # type: ignore
-        except Exception as e:
-            raise RuntimeError(
-                "llama-index-core is required when using data_dir in create_incremental_add_context"
-            ) from e
-
-        reader_kwargs: dict[str, Any] = {"recursive": True, "encoding": "utf-8"}
-        if required_exts:
-            reader_kwargs["required_exts"] = required_exts
-        documents = SimpleDirectoryReader(data_dir, **reader_kwargs).load_data(show_progress=True)
-        if documents:
-            splitter = SentenceSplitter(
-                chunk_size=chunk_size,
-                chunk_overlap=chunk_overlap,
-                separator=" ",
-                paragraph_separator="\n\n",
-            )
-            prepared_texts = []
-            for doc in documents:
-                try:
-                    nodes = splitter.get_nodes_from_documents([doc])
-                    if nodes:
-                        prepared_texts.extend([node.get_content() for node in nodes])
-                except Exception:
-                    content = doc.get_content()
-                    if content and content.strip():
-                        prepared_texts.append(content.strip())
-            if max_items > 0 and len(prepared_texts) > max_items:
-                prepared_texts = prepared_texts[:max_items]
-
-    return IncrementalAddContext(
-        index_path=index_path,
-        passages_file=passages_file,
-        offsets_file=offsets_file,
-        vector_index_file=vector_index_file,
-        embedding_model=model_name,
-        embedding_mode=mode_name,
-        distance_metric=distance_metric,
-        prepared_texts=prepared_texts,
-    )
-
-
-def incremental_add_texts_with_context(
-    ctx: IncrementalAddContext,
-    texts: list[str],
-    *,
-    ef_construction: Optional[int] = None,
-    recompute: bool = False,
-) -> int:
-    """Incrementally add texts using a prepared context (no repeated validation).
-
-    For non-compact HNSW, ef_construction (efConstruction) can be overridden during insertion.
-    """
-    if not texts:
-        return 0
-
-    # Append passages & offsets
-    _append_passages_and_update_offsets(ctx.passages_file, ctx.offsets_file, texts)
-
-    # Compute embeddings
-    # Embedding computation path
-    esm = None
-    port = None
-    if recompute:
-        passages_source_file = f"{ctx.index_path}.meta.json"
-        esm = EmbeddingServerManager(
-            backend_module_name="leann_backend_hnsw.hnsw_embedding_server",
-        )
-        started, port = esm.start_server(
-            port=5557,
-            model_name=ctx.embedding_model,
-            embedding_mode=ctx.embedding_mode,
-            passages_file=passages_source_file,
-            distance_metric=ctx.distance_metric,
-            enable_warmup=False,
-        )
-        if not started:
-            raise RuntimeError("Failed to start embedding server for recompute add")
-        embeddings = compute_embeddings_via_server(texts, ctx.embedding_model, port)
-    else:
-        embeddings = compute_embeddings(
-            texts,
-            model_name=ctx.embedding_model,
-            mode=ctx.embedding_mode,
-            use_server=False,
-            is_build=True,
-        )
-
-    # Normalize for cosine if needed
-    if ctx.distance_metric == "cosine":
-        norms = np.linalg.norm(embeddings, axis=1, keepdims=True)
-        norms[norms == 0] = 1
-        embeddings = embeddings / norms
-
-    # Append via backend helper (supports ef_construction/recompute plumbing)
-    try:
-        from leann_backend_hnsw.hnsw_backend import add_vectors as hnsw_add_vectors  # type: ignore
-    except Exception as e:
-        raise RuntimeError(
-            "Failed to import HNSW backend add helper. Ensure HNSW backend is installed."
-        ) from e
-
-    if recompute and port is not None:
-        os.environ["LEANN_ZMQ_PORT"] = str(port)
-
-    hnsw_add_vectors(
-        str(ctx.vector_index_file),
-        embeddings,
-        ef_construction=ef_construction,
-        recompute=recompute,
-    )
-
-    # Stop server after add when recompute path used
-    if esm is not None:
-        try:
-            esm.stop_server()
-        except Exception:
-            pass
-
-    return embeddings.shape[0]
-
-
-def incremental_add_directory(
-    index_path: str,
-    data_dir: str,
-    *,
-    chunk_size: int = 256,
-    chunk_overlap: int = 128,
-    required_exts: Optional[list[str]] = None,
-    max_items: int = -1,
-    embedding_model: Optional[str] = None,
-    embedding_mode: Optional[str] = None,
-) -> int:
-    """Load documents from a directory, chunk them, and incrementally add to an index.
-
-    Chunking uses LlamaIndex SentenceSplitter for simplicity and avoids external app dependencies.
-    """
-    try:
-        from llama_index.core import SimpleDirectoryReader  # type: ignore
-        from llama_index.core.node_parser import SentenceSplitter  # type: ignore
-    except Exception as e:
-        raise RuntimeError("llama-index-core is required for incremental_add_directory") from e
-
-    reader_kwargs: dict[str, Any] = {"recursive": True, "encoding": "utf-8"}
-    if required_exts:
-        reader_kwargs["required_exts"] = required_exts
-    documents = SimpleDirectoryReader(data_dir, **reader_kwargs).load_data(show_progress=True)
-    if not documents:
-        return 0
-
-    # Traditional text chunking
-    splitter = SentenceSplitter(
-        chunk_size=chunk_size,
-        chunk_overlap=chunk_overlap,
-        separator=" ",
-        paragraph_separator="\n\n",
-    )
-    all_texts: list[str] = []
-    for doc in documents:
-        try:
-            nodes = splitter.get_nodes_from_documents([doc])
-            if nodes:
-                all_texts.extend([node.get_content() for node in nodes])
-        except Exception:
-            content = doc.get_content()
-            if content and content.strip():
-                all_texts.append(content.strip())
-
-    if max_items > 0 and len(all_texts) > max_items:
-        all_texts = all_texts[:max_items]
-
-    return incremental_add_texts(
-        index_path,
-        all_texts,
-        embedding_model=embedding_model,
-        embedding_mode=embedding_mode,
-    )

packages/leann-core/src/leann/chat.py

@@ -12,6 +12,8 @@ from typing import Any, Optional
 
 import torch
 
+from .settings import resolve_ollama_host, resolve_openai_api_key, resolve_openai_base_url
+
 # Configure logging
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
@@ -310,11 +312,12 @@ def search_hf_models(query: str, limit: int = 10) -> list[str]:
 
 
 def validate_model_and_suggest(
-    model_name: str, llm_type: str, host: str = "http://localhost:11434"
+    model_name: str, llm_type: str, host: Optional[str] = None
 ) -> Optional[str]:
     """Validate model name and provide suggestions if invalid"""
     if llm_type == "ollama":
-        available_models = check_ollama_models(host)
+        resolved_host = resolve_ollama_host(host)
+        available_models = check_ollama_models(resolved_host)
         if available_models and model_name not in available_models:
             error_msg = f"Model '{model_name}' not found in your local Ollama installation."
 
@@ -457,19 +460,19 @@ class LLMInterface(ABC):
 class OllamaChat(LLMInterface):
     """LLM interface for Ollama models."""
 
-    def __init__(self, model: str = "llama3:8b", host: str = "http://localhost:11434"):
+    def __init__(self, model: str = "llama3:8b", host: Optional[str] = None):
         self.model = model
-        self.host = host
-        logger.info(f"Initializing OllamaChat with model='{model}' and host='{host}'")
+        self.host = resolve_ollama_host(host)
+        logger.info(f"Initializing OllamaChat with model='{model}' and host='{self.host}'")
         try:
             import requests
 
             # Check if the Ollama server is responsive
-            if host:
-                requests.get(host)
+            if self.host:
+                requests.get(self.host)
 
             # Pre-check model availability with helpful suggestions
-            model_error = validate_model_and_suggest(model, "ollama", host)
+            model_error = validate_model_and_suggest(model, "ollama", self.host)
             if model_error:
                 raise ValueError(model_error)
 
@@ -478,9 +481,11 @@ class OllamaChat(LLMInterface):
                 "The 'requests' library is required for Ollama. Please install it with 'pip install requests'."
             )
         except requests.exceptions.ConnectionError:
-            logger.error(f"Could not connect to Ollama at {host}. Please ensure Ollama is running.")
+            logger.error(
+                f"Could not connect to Ollama at {self.host}. Please ensure Ollama is running."
+            )
             raise ConnectionError(
-                f"Could not connect to Ollama at {host}. Please ensure Ollama is running."
+                f"Could not connect to Ollama at {self.host}. Please ensure Ollama is running."
             )
 
     def ask(self, prompt: str, **kwargs) -> str:
@@ -737,21 +742,31 @@ class GeminiChat(LLMInterface):
 class OpenAIChat(LLMInterface):
     """LLM interface for OpenAI models."""
 
-    def __init__(self, model: str = "gpt-4o", api_key: Optional[str] = None):
+    def __init__(
+        self,
+        model: str = "gpt-4o",
+        api_key: Optional[str] = None,
+        base_url: Optional[str] = None,
+    ):
         self.model = model
-        self.api_key = api_key or os.getenv("OPENAI_API_KEY")
+        self.base_url = resolve_openai_base_url(base_url)
+        self.api_key = resolve_openai_api_key(api_key)
 
         if not self.api_key:
             raise ValueError(
                 "OpenAI API key is required. Set OPENAI_API_KEY environment variable or pass api_key parameter."
             )
 
-        logger.info(f"Initializing OpenAI Chat with model='{model}'")
+        logger.info(
+            "Initializing OpenAI Chat with model='%s' and base_url='%s'",
+            model,
+            self.base_url,
+        )
 
         try:
             import openai
 
-            self.client = openai.OpenAI(api_key=self.api_key)
+            self.client = openai.OpenAI(api_key=self.api_key, base_url=self.base_url)
         except ImportError:
             raise ImportError(
                 "The 'openai' library is required for OpenAI models. Please install it with 'pip install openai'."
@@ -841,12 +856,16 @@ def get_llm(llm_config: Optional[dict[str, Any]] = None) -> LLMInterface:
     if llm_type == "ollama":
         return OllamaChat(
             model=model or "llama3:8b",
-            host=llm_config.get("host", "http://localhost:11434"),
+            host=llm_config.get("host"),
         )
     elif llm_type == "hf":
         return HFChat(model_name=model or "deepseek-ai/deepseek-llm-7b-chat")
     elif llm_type == "openai":
-        return OpenAIChat(model=model or "gpt-4o", api_key=llm_config.get("api_key"))
+        return OpenAIChat(
+            model=model or "gpt-4o",
+            api_key=llm_config.get("api_key"),
+            base_url=llm_config.get("base_url"),
+        )
     elif llm_type == "gemini":
         return GeminiChat(model=model or "gemini-2.5-flash", api_key=llm_config.get("api_key"))
     elif llm_type == "simulated":

packages/leann-core/src/leann/cli.py

@@ -9,6 +9,7 @@ from tqdm import tqdm
 
 from .api import LeannBuilder, LeannChat, LeannSearcher
 from .registry import register_project_directory
+from .settings import resolve_ollama_host, resolve_openai_api_key, resolve_openai_base_url
 
 
 def extract_pdf_text_with_pymupdf(file_path: str) -> str:
@@ -123,6 +124,24 @@ Examples:
             choices=["sentence-transformers", "openai", "mlx", "ollama"],
             help="Embedding backend mode (default: sentence-transformers)",
         )
+        build_parser.add_argument(
+            "--embedding-host",
+            type=str,
+            default=None,
+            help="Override Ollama-compatible embedding host",
+        )
+        build_parser.add_argument(
+            "--embedding-api-base",
+            type=str,
+            default=None,
+            help="Base URL for OpenAI-compatible embedding services",
+        )
+        build_parser.add_argument(
+            "--embedding-api-key",
+            type=str,
+            default=None,
+            help="API key for embedding service (defaults to OPENAI_API_KEY)",
+        )
         build_parser.add_argument(
             "--force", "-f", action="store_true", help="Force rebuild existing index"
         )
@@ -248,7 +267,12 @@ Examples:
         ask_parser.add_argument(
             "--model", type=str, default="qwen3:8b", help="Model name (default: qwen3:8b)"
         )
-        ask_parser.add_argument("--host", type=str, default="http://localhost:11434")
+        ask_parser.add_argument(
+            "--host",
+            type=str,
+            default=None,
+            help="Override Ollama-compatible host (defaults to LEANN_OLLAMA_HOST/OLLAMA_HOST)",
+        )
         ask_parser.add_argument(
             "--interactive", "-i", action="store_true", help="Interactive chat mode"
         )
@@ -277,6 +301,18 @@ Examples:
             default=None,
             help="Thinking budget for reasoning models (low/medium/high). Supported by GPT-Oss:20b and other reasoning models.",
         )
+        ask_parser.add_argument(
+            "--api-base",
+            type=str,
+            default=None,
+            help="Base URL for OpenAI-compatible APIs (e.g., http://localhost:10000/v1)",
+        )
+        ask_parser.add_argument(
+            "--api-key",
+            type=str,
+            default=None,
+            help="API key for OpenAI-compatible APIs (defaults to OPENAI_API_KEY)",
+        )
 
         # List command
         subparsers.add_parser("list", help="List all indexes")
@@ -1325,10 +1361,20 @@ Examples:
 
         print(f"Building index '{index_name}' with {args.backend} backend...")
 
+        embedding_options: dict[str, Any] = {}
+        if args.embedding_mode == "ollama":
+            embedding_options["host"] = resolve_ollama_host(args.embedding_host)
+        elif args.embedding_mode == "openai":
+            embedding_options["base_url"] = resolve_openai_base_url(args.embedding_api_base)
+            resolved_embedding_key = resolve_openai_api_key(args.embedding_api_key)
+            if resolved_embedding_key:
+                embedding_options["api_key"] = resolved_embedding_key
+
         builder = LeannBuilder(
             backend_name=args.backend,
             embedding_model=args.embedding_model,
             embedding_mode=args.embedding_mode,
+            embedding_options=embedding_options or None,
             graph_degree=args.graph_degree,
             complexity=args.complexity,
             is_compact=args.compact,
@@ -1476,7 +1522,12 @@ Examples:
 
         llm_config = {"type": args.llm, "model": args.model}
         if args.llm == "ollama":
-            llm_config["host"] = args.host
+            llm_config["host"] = resolve_ollama_host(args.host)
+        elif args.llm == "openai":
+            llm_config["base_url"] = resolve_openai_base_url(args.api_base)
+            resolved_api_key = resolve_openai_api_key(args.api_key)
+            if resolved_api_key:
+                llm_config["api_key"] = resolved_api_key
 
         chat = LeannChat(index_path=index_path, llm_config=llm_config)
 

packages/leann-core/src/leann/embedding_compute.py

@@ -7,11 +7,13 @@ Preserves all optimization parameters to ensure performance
 import logging
 import os
 import time
-from typing import Any
+from typing import Any, Optional
 
 import numpy as np
 import torch
 
+from .settings import resolve_ollama_host, resolve_openai_api_key, resolve_openai_base_url
+
 # Set up logger with proper level
 logger = logging.getLogger(__name__)
 LOG_LEVEL = os.getenv("LEANN_LOG_LEVEL", "WARNING").upper()
@@ -31,6 +33,7 @@ def compute_embeddings(
     adaptive_optimization: bool = True,
     manual_tokenize: bool = False,
     max_length: int = 512,
+    provider_options: Optional[dict[str, Any]] = None,
 ) -> np.ndarray:
     """
     Unified embedding computation entry point
@@ -46,6 +49,8 @@ def compute_embeddings(
     Returns:
         Normalized embeddings array, shape: (len(texts), embedding_dim)
     """
+    provider_options = provider_options or {}
+
     if mode == "sentence-transformers":
         return compute_embeddings_sentence_transformers(
             texts,
@@ -57,11 +62,21 @@ def compute_embeddings(
             max_length=max_length,
         )
     elif mode == "openai":
-        return compute_embeddings_openai(texts, model_name)
+        return compute_embeddings_openai(
+            texts,
+            model_name,
+            base_url=provider_options.get("base_url"),
+            api_key=provider_options.get("api_key"),
+        )
     elif mode == "mlx":
         return compute_embeddings_mlx(texts, model_name)
     elif mode == "ollama":
-        return compute_embeddings_ollama(texts, model_name, is_build=is_build)
+        return compute_embeddings_ollama(
+            texts,
+            model_name,
+            is_build=is_build,
+            host=provider_options.get("host"),
+        )
     elif mode == "gemini":
         return compute_embeddings_gemini(texts, model_name, is_build=is_build)
     else:
@@ -353,12 +368,15 @@ def compute_embeddings_sentence_transformers(
     return embeddings
 
 
-def compute_embeddings_openai(texts: list[str], model_name: str) -> np.ndarray:
+def compute_embeddings_openai(
+    texts: list[str],
+    model_name: str,
+    base_url: Optional[str] = None,
+    api_key: Optional[str] = None,
+) -> np.ndarray:
     # TODO: @yichuan-w add progress bar only in build mode
     """Compute embeddings using OpenAI API"""
     try:
-        import os
-
         import openai
     except ImportError as e:
         raise ImportError(f"OpenAI package not installed: {e}")
@@ -373,16 +391,18 @@ def compute_embeddings_openai(texts: list[str], model_name: str) -> np.ndarray:
             f"Found {invalid_count} empty/invalid text(s) in input. Upstream should filter before calling OpenAI."
         )
 
-    api_key = os.getenv("OPENAI_API_KEY")
-    if not api_key:
+    resolved_base_url = resolve_openai_base_url(base_url)
+    resolved_api_key = resolve_openai_api_key(api_key)
+
+    if not resolved_api_key:
         raise RuntimeError("OPENAI_API_KEY environment variable not set")
 
     # Cache OpenAI client
-    cache_key = "openai_client"
+    cache_key = f"openai_client::{resolved_base_url}"
     if cache_key in _model_cache:
         client = _model_cache[cache_key]
     else:
-        client = openai.OpenAI(api_key=api_key)
+        client = openai.OpenAI(api_key=resolved_api_key, base_url=resolved_base_url)
         _model_cache[cache_key] = client
         logger.info("OpenAI client cached")
 
@@ -507,7 +527,10 @@ def compute_embeddings_mlx(chunks: list[str], model_name: str, batch_size: int =
 
 
 def compute_embeddings_ollama(
-    texts: list[str], model_name: str, is_build: bool = False, host: str = "http://localhost:11434"
+    texts: list[str],
+    model_name: str,
+    is_build: bool = False,
+    host: Optional[str] = None,
 ) -> np.ndarray:
     """
     Compute embeddings using Ollama API with simplified batch processing.
@@ -518,7 +541,7 @@ def compute_embeddings_ollama(
         texts: List of texts to compute embeddings for
         model_name: Ollama model name (e.g., "nomic-embed-text", "mxbai-embed-large")
         is_build: Whether this is a build operation (shows progress bar)
-        host: Ollama host URL (default: http://localhost:11434)
+        host: Ollama host URL (defaults to environment or http://localhost:11434)
 
     Returns:
         Normalized embeddings array, shape: (len(texts), embedding_dim)
@@ -533,17 +556,19 @@ def compute_embeddings_ollama(
     if not texts:
         raise ValueError("Cannot compute embeddings for empty text list")
 
+    resolved_host = resolve_ollama_host(host)
+
     logger.info(
-        f"Computing embeddings for {len(texts)} texts using Ollama API, model: '{model_name}'"
+        f"Computing embeddings for {len(texts)} texts using Ollama API, model: '{model_name}', host: '{resolved_host}'"
     )
 
     # Check if Ollama is running
     try:
-        response = requests.get(f"{host}/api/version", timeout=5)
+        response = requests.get(f"{resolved_host}/api/version", timeout=5)
         response.raise_for_status()
     except requests.exceptions.ConnectionError:
         error_msg = (
-            f"❌ Could not connect to Ollama at {host}.\n\n"
+            f"❌ Could not connect to Ollama at {resolved_host}.\n\n"
             "Please ensure Ollama is running:\n"
             "  • macOS/Linux: ollama serve\n"
             "  • Windows: Make sure Ollama is running in the system tray\n\n"
@@ -555,7 +580,7 @@ def compute_embeddings_ollama(
 
     # Check if model exists and provide helpful suggestions
     try:
-        response = requests.get(f"{host}/api/tags", timeout=5)
+        response = requests.get(f"{resolved_host}/api/tags", timeout=5)
         response.raise_for_status()
         models = response.json()
         model_names = [model["name"] for model in models.get("models", [])]
@@ -618,7 +643,9 @@ def compute_embeddings_ollama(
         # Verify the model supports embeddings by testing it
         try:
             test_response = requests.post(
-                f"{host}/api/embeddings", json={"model": model_name, "prompt": "test"}, timeout=10
+                f"{resolved_host}/api/embeddings",
+                json={"model": model_name, "prompt": "test"},
+                timeout=10,
             )
             if test_response.status_code != 200:
                 error_msg = (
@@ -665,7 +692,7 @@ def compute_embeddings_ollama(
             while retry_count < max_retries:
                 try:
                     response = requests.post(
-                        f"{host}/api/embeddings",
+                        f"{resolved_host}/api/embeddings",
                         json={"model": model_name, "prompt": truncated_text},
                         timeout=30,
                     )

packages/leann-core/src/leann/embedding_server_manager.py

@@ -8,6 +8,8 @@ import time
 from pathlib import Path
 from typing import Optional
 
+from .settings import encode_provider_options
+
 # Lightweight, self-contained server manager with no cross-process inspection
 
 # Set up logging based on environment variable
@@ -82,16 +84,40 @@ class EmbeddingServerManager:
     ) -> tuple[bool, int]:
         """Start the embedding server."""
         # passages_file may be present in kwargs for server CLI, but we don't need it here
+        provider_options = kwargs.pop("provider_options", None)
+
+        config_signature = {
+            "model_name": model_name,
+            "passages_file": kwargs.get("passages_file", ""),
+            "embedding_mode": embedding_mode,
+            "provider_options": provider_options or {},
+        }
 
         # If this manager already has a live server, just reuse it
-        if self.server_process and self.server_process.poll() is None and self.server_port:
+        if (
+            self.server_process
+            and self.server_process.poll() is None
+            and self.server_port
+            and self._server_config == config_signature
+        ):
             logger.info("Reusing in-process server")
             return True, self.server_port
 
+        # Configuration changed, stop existing server before starting a new one
+        if self.server_process and self.server_process.poll() is None:
+            logger.info("Existing server configuration differs; restarting embedding server")
+            self.stop_server()
+
         # For Colab environment, use a different strategy
         if _is_colab_environment():
             logger.info("Detected Colab environment, using alternative startup strategy")
-            return self._start_server_colab(port, model_name, embedding_mode, **kwargs)
+            return self._start_server_colab(
+                port,
+                model_name,
+                embedding_mode,
+                provider_options=provider_options,
+                **kwargs,
+            )
 
         # Always pick a fresh available port
         try:
@@ -101,13 +127,21 @@ class EmbeddingServerManager:
             return False, port
 
         # Start a new server
-        return self._start_new_server(actual_port, model_name, embedding_mode, **kwargs)
+        return self._start_new_server(
+            actual_port,
+            model_name,
+            embedding_mode,
+            provider_options=provider_options,
+            config_signature=config_signature,
+            **kwargs,
+        )
 
     def _start_server_colab(
         self,
         port: int,
         model_name: str,
         embedding_mode: str = "sentence-transformers",
+        provider_options: Optional[dict] = None,
         **kwargs,
     ) -> tuple[bool, int]:
         """Start server with Colab-specific configuration."""
@@ -125,8 +159,20 @@ class EmbeddingServerManager:
 
         try:
             # In Colab, we'll use a more direct approach
-            self._launch_server_process_colab(command, actual_port)
-            return self._wait_for_server_ready_colab(actual_port)
+            self._launch_server_process_colab(
+                command,
+                actual_port,
+                provider_options=provider_options,
+            )
+            started, ready_port = self._wait_for_server_ready_colab(actual_port)
+            if started:
+                self._server_config = {
+                    "model_name": model_name,
+                    "passages_file": kwargs.get("passages_file", ""),
+                    "embedding_mode": embedding_mode,
+                    "provider_options": provider_options or {},
+                }
+            return started, ready_port
         except Exception as e:
             logger.error(f"Failed to start embedding server in Colab: {e}")
             return False, actual_port
@@ -134,7 +180,13 @@ class EmbeddingServerManager:
     # Note: No compatibility check needed; manager is per-searcher and configs are stable per instance
 
     def _start_new_server(
-        self, port: int, model_name: str, embedding_mode: str, **kwargs
+        self,
+        port: int,
+        model_name: str,
+        embedding_mode: str,
+        provider_options: Optional[dict] = None,
+        config_signature: Optional[dict] = None,
+        **kwargs,
     ) -> tuple[bool, int]:
         """Start a new embedding server on the given port."""
         logger.info(f"Starting embedding server on port {port}...")
@@ -142,8 +194,20 @@ class EmbeddingServerManager:
         command = self._build_server_command(port, model_name, embedding_mode, **kwargs)
 
         try:
-            self._launch_server_process(command, port)
-            return self._wait_for_server_ready(port)
+            self._launch_server_process(
+                command,
+                port,
+                provider_options=provider_options,
+            )
+            started, ready_port = self._wait_for_server_ready(port)
+            if started:
+                self._server_config = config_signature or {
+                    "model_name": model_name,
+                    "passages_file": kwargs.get("passages_file", ""),
+                    "embedding_mode": embedding_mode,
+                    "provider_options": provider_options or {},
+                }
+            return started, ready_port
         except Exception as e:
             logger.error(f"Failed to start embedding server: {e}")
             return False, port
@@ -173,7 +237,12 @@ class EmbeddingServerManager:
 
         return command
 
-    def _launch_server_process(self, command: list, port: int) -> None:
+    def _launch_server_process(
+        self,
+        command: list,
+        port: int,
+        provider_options: Optional[dict] = None,
+    ) -> None:
         """Launch the server process."""
         project_root = Path(__file__).parent.parent.parent.parent.parent
         logger.info(f"Command: {' '.join(command)}")
@@ -193,14 +262,20 @@ class EmbeddingServerManager:
 
         # Start embedding server subprocess
         logger.info(f"Starting server process with command: {' '.join(command)}")
+        env = os.environ.copy()
+        encoded_options = encode_provider_options(provider_options)
+        if encoded_options:
+            env["LEANN_EMBEDDING_OPTIONS"] = encoded_options
+
         self.server_process = subprocess.Popen(
             command,
             cwd=project_root,
             stdout=stdout_target,
             stderr=stderr_target,
+            env=env,
         )
         self.server_port = port
-        # Record config for in-process reuse
+        # Record config for in-process reuse (best effort; refined later when ready)
         try:
             self._server_config = {
                 "model_name": command[command.index("--model-name") + 1]
@@ -212,12 +287,14 @@ class EmbeddingServerManager:
                 "embedding_mode": command[command.index("--embedding-mode") + 1]
                 if "--embedding-mode" in command
                 else "sentence-transformers",
+                "provider_options": provider_options or {},
             }
         except Exception:
             self._server_config = {
                 "model_name": "",
                 "passages_file": "",
                 "embedding_mode": "sentence-transformers",
+                "provider_options": provider_options or {},
             }
         logger.info(f"Server process started with PID: {self.server_process.pid}")
 
@@ -322,16 +399,27 @@ class EmbeddingServerManager:
         # Removed: cross-process adoption no longer supported
         return
 
-    def _launch_server_process_colab(self, command: list, port: int) -> None:
+    def _launch_server_process_colab(
+        self,
+        command: list,
+        port: int,
+        provider_options: Optional[dict] = None,
+    ) -> None:
         """Launch the server process with Colab-specific settings."""
         logger.info(f"Colab Command: {' '.join(command)}")
 
         # In Colab, we need to be more careful about process management
+        env = os.environ.copy()
+        encoded_options = encode_provider_options(provider_options)
+        if encoded_options:
+            env["LEANN_EMBEDDING_OPTIONS"] = encoded_options
+
         self.server_process = subprocess.Popen(
             command,
             stdout=subprocess.PIPE,
             stderr=subprocess.PIPE,
             text=True,
+            env=env,
         )
         self.server_port = port
         logger.info(f"Colab server process started with PID: {self.server_process.pid}")
@@ -345,6 +433,7 @@ class EmbeddingServerManager:
             "model_name": "",
             "passages_file": "",
             "embedding_mode": "sentence-transformers",
+            "provider_options": provider_options or {},
         }
 
     def _wait_for_server_ready_colab(self, port: int) -> tuple[bool, int]:

packages/leann-core/src/leann/searcher_base.py

@@ -41,6 +41,7 @@ class BaseSearcher(LeannBackendSearcherInterface, ABC):
             print("WARNING: embedding_model not found in meta.json. Recompute will fail.")
 
         self.embedding_mode = self.meta.get("embedding_mode", "sentence-transformers")
+        self.embedding_options = self.meta.get("embedding_options", {})
 
         self.embedding_server_manager = EmbeddingServerManager(
             backend_module_name=backend_module_name,
@@ -77,6 +78,7 @@ class BaseSearcher(LeannBackendSearcherInterface, ABC):
             passages_file=passages_source_file,
             distance_metric=distance_metric,
             enable_warmup=kwargs.get("enable_warmup", False),
+            provider_options=self.embedding_options,
         )
         if not server_started:
             raise RuntimeError(f"Failed to start embedding server on port {actual_port}")
@@ -125,7 +127,12 @@ class BaseSearcher(LeannBackendSearcherInterface, ABC):
         from .embedding_compute import compute_embeddings
 
         embedding_mode = self.meta.get("embedding_mode", "sentence-transformers")
-        return compute_embeddings([query], self.embedding_model, embedding_mode)
+        return compute_embeddings(
+            [query],
+            self.embedding_model,
+            embedding_mode,
+            provider_options=self.embedding_options,
+        )
 
     def _compute_embedding_via_server(self, chunks: list, zmq_port: int) -> np.ndarray:
         """Compute embeddings using the ZMQ embedding server."""

packages/leann-core/src/leann/settings.py

@@ -0,0 +1,74 @@
+"""Runtime configuration helpers for LEANN."""
+
+from __future__ import annotations
+
+import json
+import os
+from typing import Any
+
+# Default fallbacks to preserve current behaviour while keeping them in one place.
+_DEFAULT_OLLAMA_HOST = "http://localhost:11434"
+_DEFAULT_OPENAI_BASE_URL = "https://api.openai.com/v1"
+
+
+def _clean_url(value: str) -> str:
+    """Normalize URL strings by stripping trailing slashes."""
+
+    return value.rstrip("/") if value else value
+
+
+def resolve_ollama_host(explicit: str | None = None) -> str:
+    """Resolve the Ollama-compatible endpoint to use."""
+
+    candidates = (
+        explicit,
+        os.getenv("LEANN_LOCAL_LLM_HOST"),
+        os.getenv("LEANN_OLLAMA_HOST"),
+        os.getenv("OLLAMA_HOST"),
+        os.getenv("LOCAL_LLM_ENDPOINT"),
+    )
+
+    for candidate in candidates:
+        if candidate:
+            return _clean_url(candidate)
+
+    return _clean_url(_DEFAULT_OLLAMA_HOST)
+
+
+def resolve_openai_base_url(explicit: str | None = None) -> str:
+    """Resolve the base URL for OpenAI-compatible services."""
+
+    candidates = (
+        explicit,
+        os.getenv("LEANN_OPENAI_BASE_URL"),
+        os.getenv("OPENAI_BASE_URL"),
+        os.getenv("LOCAL_OPENAI_BASE_URL"),
+    )
+
+    for candidate in candidates:
+        if candidate:
+            return _clean_url(candidate)
+
+    return _clean_url(_DEFAULT_OPENAI_BASE_URL)
+
+
+def resolve_openai_api_key(explicit: str | None = None) -> str | None:
+    """Resolve the API key for OpenAI-compatible services."""
+
+    if explicit:
+        return explicit
+
+    return os.getenv("OPENAI_API_KEY")
+
+
+def encode_provider_options(options: dict[str, Any] | None) -> str | None:
+    """Serialize provider options for child processes."""
+
+    if not options:
+        return None
+
+    try:
+        return json.dumps(options)
+    except (TypeError, ValueError):
+        # Fall back to empty payload if serialization fails
+        return None