Add ty type checker to CI and fix type errors (fixes bug from PR #157) (#192)

* Add ty type checker to CI and fix type errors

- Add ty (Astral's fast Python type checker) to GitHub CI workflow
- Fix type annotations across all RAG apps:
  - Update load_data return types from list[str] to list[dict[str, Any]]
  - Fix base_rag_example.py to properly handle dict format from create_text_chunks
- Fix type errors in leann-core:
  - chunking_utils.py: Add explicit type annotations
  - cli.py: Fix return type annotations for PDF extraction functions
  - interactive_utils.py: Fix readline import type handling
- Fix type errors in apps:
  - wechat_history.py: Fix return type annotations
  - document_rag.py, code_rag.py: Replace **kwargs with explicit arguments
- Add ty configuration to pyproject.toml

This resolves the bug introduced in PR #157 where create_text_chunks()
changed to return list[dict] but callers were not updated.

* Fix remaining ty type errors

- Fix slack_mcp_reader.py channel parameter can be None
- Fix embedding_compute.py ContextProp type issue
- Fix searcher_base.py method override signatures
- Fix chunking_utils.py chunk_text assignment
- Fix slack_rag.py and twitter_rag.py return types
- Fix email.py and image_rag.py method overrides

* Fix multimodal benchmark scripts type errors

- Fix undefined LeannRetriever -> LeannMultiVector
- Add proper type casts for HuggingFace Dataset iteration
- Cast task config values to correct types
- Add type annotations for dataset row dicts

* Enable ty check for multimodal scripts in CI

All type errors in multimodal scripts have been fixed, so we can now
include them in the CI type checking.

* Fix all test type errors and enable ty check on tests

- Fix test_basic.py: search() takes str not list
- Fix test_cli_prompt_template.py: add type: ignore for Mock assignments
- Fix test_prompt_template_persistence.py: match BaseSearcher.search signature
- Fix test_prompt_template_e2e.py: add type narrowing asserts after skip
- Fix test_readme_examples.py: use explicit kwargs instead of **model_args
- Fix metadata_filter.py: allow Optional[MetadataFilters]
- Update CI to run ty check on tests

* Format code with ruff

* Format searcher_base.py

.github/workflows/build-reusable.yml

@@ -28,15 +28,36 @@ jobs:
         run: |
           uv run --only-group lint pre-commit run --all-files --show-diff-on-failure
 
+  type-check:
+    name: Type Check with ty
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.ref }}
+          submodules: recursive
+
+      - name: Install uv and Python
+        uses: astral-sh/setup-uv@v6
+        with:
+          python-version: '3.11'
+
+      - name: Install ty
+        run: uv tool install ty
+
+      - name: Run ty type checker
+        run: |
+          # Run ty on core packages, apps, and tests
+          ty check packages/leann-core/src apps tests
 
   build:
-    needs: lint
+    needs: [lint, type-check]
     name: Build ${{ matrix.os }} Python ${{ matrix.python }}
     strategy:
       matrix:
         include:
-          - os: ubuntu-22.04
+          # Note: Python 3.9 dropped - uses PEP 604 union syntax (str | None)
-            python: '3.9'
+          # which requires Python 3.10+
           - os: ubuntu-22.04
             python: '3.10'
           - os: ubuntu-22.04
@@ -46,8 +67,6 @@ jobs:
           - os: ubuntu-22.04
             python: '3.13'
           # ARM64 Linux builds
-          - os: ubuntu-24.04-arm
-            python: '3.9'
           - os: ubuntu-24.04-arm
             python: '3.10'
           - os: ubuntu-24.04-arm
@@ -56,8 +75,6 @@ jobs:
             python: '3.12'
           - os: ubuntu-24.04-arm
             python: '3.13'
-          - os: macos-14
-            python: '3.9'
           - os: macos-14
             python: '3.10'
           - os: macos-14
@@ -66,8 +83,6 @@ jobs:
             python: '3.12'
           - os: macos-14
             python: '3.13'
-          - os: macos-15
-            python: '3.9'
           - os: macos-15
             python: '3.10'
           - os: macos-15
@@ -76,16 +91,24 @@ jobs:
             python: '3.12'
           - os: macos-15
             python: '3.13'
-          - os: macos-13
+          # Intel Mac builds (x86_64) - replaces deprecated macos-13
-            python: '3.9'
+          # Note: Python 3.13 excluded - PyTorch has no wheels for macOS x86_64 + Python 3.13
-          - os: macos-13
+          # (PyTorch <=2.4.1 lacks cp313, PyTorch >=2.5.0 dropped Intel Mac support)
+          - os: macos-15-intel
             python: '3.10'
-          - os: macos-13
+          - os: macos-15-intel
             python: '3.11'
-          - os: macos-13
+          - os: macos-15-intel
             python: '3.12'
-          # Note: macos-13 + Python 3.13 excluded due to PyTorch compatibility
+          # macOS 26 (beta) - arm64
-          # (PyTorch 2.5+ supports Python 3.13 but not Intel Mac x86_64)
+          - os: macos-26
+            python: '3.10'
+          - os: macos-26
+            python: '3.11'
+          - os: macos-26
+            python: '3.12'
+          - os: macos-26
+            python: '3.13'
     runs-on: ${{ matrix.os }}
 
     steps:
@@ -204,13 +227,16 @@ jobs:
             # Use system clang for better compatibility
             export CC=clang
             export CXX=clang++
-            # Homebrew libraries on each macOS version require matching minimum version
+            # Set deployment target based on runner
-            if [[ "${{ matrix.os }}" == "macos-13" ]]; then
+            # macos-15-intel runs macOS 15, so target 15.0 (system libraries require it)
-              export MACOSX_DEPLOYMENT_TARGET=13.0
+            if [[ "${{ matrix.os }}" == "macos-15-intel" ]]; then
-            elif [[ "${{ matrix.os }}" == "macos-14" ]]; then
-              export MACOSX_DEPLOYMENT_TARGET=14.0
-            elif [[ "${{ matrix.os }}" == "macos-15" ]]; then
               export MACOSX_DEPLOYMENT_TARGET=15.0
+            elif [[ "${{ matrix.os }}" == macos-14* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=14.0
+            elif [[ "${{ matrix.os }}" == macos-15* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=15.0
+            elif [[ "${{ matrix.os }}" == macos-26* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=26.0
             fi
             uv build --wheel --python ${{ matrix.python }} --find-links ${GITHUB_WORKSPACE}/packages/leann-core/dist
           else
@@ -224,14 +250,16 @@ jobs:
             # Use system clang for better compatibility
             export CC=clang
             export CXX=clang++
-            # DiskANN requires macOS 13.3+ for sgesdd_ LAPACK function
+            # Set deployment target based on runner
-            # But Homebrew libraries on each macOS version require matching minimum version
+            # macos-15-intel runs macOS 15, so target 15.0 (system libraries require it)
-            if [[ "${{ matrix.os }}" == "macos-13" ]]; then
+            if [[ "${{ matrix.os }}" == "macos-15-intel" ]]; then
-              export MACOSX_DEPLOYMENT_TARGET=13.3
-            elif [[ "${{ matrix.os }}" == "macos-14" ]]; then
-              export MACOSX_DEPLOYMENT_TARGET=14.0
-            elif [[ "${{ matrix.os }}" == "macos-15" ]]; then
               export MACOSX_DEPLOYMENT_TARGET=15.0
+            elif [[ "${{ matrix.os }}" == macos-14* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=14.0
+            elif [[ "${{ matrix.os }}" == macos-15* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=15.0
+            elif [[ "${{ matrix.os }}" == macos-26* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=26.0
             fi
             uv build --wheel --python ${{ matrix.python }} --find-links ${GITHUB_WORKSPACE}/packages/leann-core/dist
           else
@@ -269,16 +297,19 @@ jobs:
         if: runner.os == 'macOS'
         run: |
           # Determine deployment target based on runner OS
-          # Must match the Homebrew libraries for each macOS version
+          # macos-15-intel runs macOS 15, so target 15.0 (system libraries require it)
-          if [[ "${{ matrix.os }}" == "macos-13" ]]; then
+          if [[ "${{ matrix.os }}" == "macos-15-intel" ]]; then
-            HNSW_TARGET="13.0"
-            DISKANN_TARGET="13.3"
-          elif [[ "${{ matrix.os }}" == "macos-14" ]]; then
-            HNSW_TARGET="14.0"
-            DISKANN_TARGET="14.0"
-          elif [[ "${{ matrix.os }}" == "macos-15" ]]; then
             HNSW_TARGET="15.0"
             DISKANN_TARGET="15.0"
+          elif [[ "${{ matrix.os }}" == macos-14* ]]; then
+            HNSW_TARGET="14.0"
+            DISKANN_TARGET="14.0"
+          elif [[ "${{ matrix.os }}" == macos-15* ]]; then
+            HNSW_TARGET="15.0"
+            DISKANN_TARGET="15.0"
+          elif [[ "${{ matrix.os }}" == macos-26* ]]; then
+            HNSW_TARGET="26.0"
+            DISKANN_TARGET="26.0"
           fi
 
           # Repair HNSW wheel
@@ -334,12 +365,15 @@ jobs:
           PY_TAG=$($UV_PY -c "import sys; print(f'cp{sys.version_info[0]}{sys.version_info[1]}')")
 
           if [[ "$RUNNER_OS" == "macOS" ]]; then
-            if [[ "${{ matrix.os }}" == "macos-13" ]]; then
+            # macos-15-intel runs macOS 15, so target 15.0 (system libraries require it)
-              export MACOSX_DEPLOYMENT_TARGET=13.3
+            if [[ "${{ matrix.os }}" == "macos-15-intel" ]]; then
-            elif [[ "${{ matrix.os }}" == "macos-14" ]]; then
-              export MACOSX_DEPLOYMENT_TARGET=14.0
-            elif [[ "${{ matrix.os }}" == "macos-15" ]]; then
               export MACOSX_DEPLOYMENT_TARGET=15.0
+            elif [[ "${{ matrix.os }}" == macos-14* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=14.0
+            elif [[ "${{ matrix.os }}" == macos-15* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=15.0
+            elif [[ "${{ matrix.os }}" == macos-26* ]]; then
+              export MACOSX_DEPLOYMENT_TARGET=26.0
             fi
           fi
 

README.md

@@ -36,7 +36,7 @@ LEANN is an innovative vector database that democratizes personal AI. Transform
 
 LEANN achieves this through *graph-based selective recomputation* with *high-degree preserving pruning*, computing embeddings on-demand instead of storing them all. [Illustration Fig →](#️-architecture--how-it-works) | [Paper →](https://arxiv.org/abs/2506.08276)
 
-**Ready to RAG Everything?** Transform your laptop into a personal AI assistant that can semantic search your **[file system](#-personal-data-manager-process-any-documents-pdf-txt-md)**, **[emails](#-your-personal-email-secretary-rag-on-apple-mail)**, **[browser history](#-time-machine-for-the-web-rag-your-entire-browser-history)**, **[chat history](#-wechat-detective-unlock-your-golden-memories)** ([WeChat](#-wechat-detective-unlock-your-golden-memories), [iMessage](#-imessage-history-your-personal-conversation-archive)), **[agent memory](#-chatgpt-chat-history-your-personal-ai-conversation-archive)** ([ChatGPT](#-chatgpt-chat-history-your-personal-ai-conversation-archive), [Claude](#-claude-chat-history-your-personal-ai-conversation-archive)), **[live data](#mcp-integration-rag-on-live-data-from-any-platform)** ([Slack](#mcp-integration-rag-on-live-data-from-any-platform), [Twitter](#mcp-integration-rag-on-live-data-from-any-platform)), **[codebase](#-claude-code-integration-transform-your-development-workflow)**\* , or external knowledge bases (i.e., 60M documents) - all on your laptop, with zero cloud costs and complete privacy.
+**Ready to RAG Everything?** Transform your laptop into a personal AI assistant that can semantic search your **[file system](#-personal-data-manager-process-any-documents-pdf-txt-md)**, **[emails](#-your-personal-email-secretary-rag-on-apple-mail)**, **[browser history](#-time-machine-for-the-web-rag-your-entire-browser-history)**, **[chat history](#-wechat-detective-unlock-your-golden-memories)** ([WeChat](#-wechat-detective-unlock-your-golden-memories), [iMessage](#-imessage-history-your-personal-conversation-archive)), **[agent memory](#-chatgpt-chat-history-your-personal-ai-conversation-archive)** ([ChatGPT](#-chatgpt-chat-history-your-personal-ai-conversation-archive), [Claude](#-claude-chat-history-your-personal-ai-conversation-archive)), **[live data](#mcp-integration-rag-on-live-data-from-any-platform)** ([Slack](#slack-messages-search-your-team-conversations), [Twitter](#-twitter-bookmarks-your-personal-tweet-library)), **[codebase](#-claude-code-integration-transform-your-development-workflow)**\* , or external knowledge bases (i.e., 60M documents) - all on your laptop, with zero cloud costs and complete privacy.
 
 
 \* Claude Code only supports basic `grep`-style keyword search. **LEANN** is a drop-in **semantic search MCP service fully compatible with Claude Code**, unlocking intelligent retrieval without changing your workflow. 🔥 Check out [the easy setup →](packages/leann-mcp/README.md)
@@ -392,6 +392,54 @@ python -m apps.code_rag --repo-dir "./my_codebase" --query "How does authenticat
 
 </details>
 
+### 🎨 ColQwen: Multimodal PDF Retrieval with Vision-Language Models
+
+Search through PDFs using both text and visual understanding with ColQwen2/ColPali models. Perfect for research papers, technical documents, and any PDFs with complex layouts, figures, or diagrams.
+
+> **🍎 Mac Users**: ColQwen is optimized for Apple Silicon with MPS acceleration for faster inference!
+
+```bash
+# Build index from PDFs
+python -m apps.colqwen_rag build --pdfs ./my_papers/ --index research_papers
+
+# Search with text queries
+python -m apps.colqwen_rag search research_papers "How does attention mechanism work?"
+
+# Interactive Q&A
+python -m apps.colqwen_rag ask research_papers --interactive
+```
+
+<details>
+<summary><strong>📋 Click to expand: ColQwen Setup & Usage</strong></summary>
+
+#### Prerequisites
+```bash
+# Install dependencies
+uv pip install colpali_engine pdf2image pillow matplotlib qwen_vl_utils einops seaborn
+brew install poppler  # macOS only, for PDF processing
+```
+
+#### Build Index
+```bash
+python -m apps.colqwen_rag build \
+  --pdfs ./pdf_directory/ \
+  --index my_index \
+  --model colqwen2  # or colpali
+```
+
+#### Search
+```bash
+python -m apps.colqwen_rag search my_index "your question here" --top-k 5
+```
+
+#### Models
+- **ColQwen2** (`colqwen2`): Latest vision-language model with improved performance
+- **ColPali** (`colpali`): Proven multimodal retriever
+
+For detailed usage, see the [ColQwen Guide](docs/COLQWEN_GUIDE.md).
+
+</details>
+
 ### 📧 Your Personal Email Secretary: RAG on Apple Mail!
 
 > **Note:** The examples below currently support macOS only. Windows support coming soon.

apps/base_rag_example.py

@@ -257,8 +257,8 @@ class BaseRAGExample(ABC):
         pass
 
     @abstractmethod
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
-        """Load data from the source. Returns list of text chunks."""
+        """Load data from the source. Returns list of text chunks as dicts with 'text' and 'metadata' keys."""
         pass
 
     def get_llm_config(self, args) -> dict[str, Any]:
@@ -282,8 +282,8 @@ class BaseRAGExample(ABC):
 
         return config
 
-    async def build_index(self, args, texts: list[str]) -> str:
+    async def build_index(self, args, texts: list[dict[str, Any]]) -> str:
-        """Build LEANN index from texts."""
+        """Build LEANN index from text chunks (dicts with 'text' and 'metadata' keys)."""
         index_path = str(Path(args.index_dir) / f"{self.default_index_name}.leann")
 
         print(f"\n[Building Index] Creating {self.name} index...")
@@ -314,8 +314,14 @@ class BaseRAGExample(ABC):
         batch_size = 1000
         for i in range(0, len(texts), batch_size):
             batch = texts[i : i + batch_size]
-            for text in batch:
+            for item in batch:
-                builder.add_text(text)
+                # Handle both dict format (from create_text_chunks) and plain strings
+                if isinstance(item, dict):
+                    text = item.get("text", "")
+                    metadata = item.get("metadata")
+                    builder.add_text(text, metadata)
+                else:
+                    builder.add_text(item)
             print(f"Added {min(i + batch_size, len(texts))}/{len(texts)} texts...")
 
         print("Building index structure...")

apps/browser_rag.py

@@ -6,6 +6,7 @@ Supports Chrome browser history.
 import os
 import sys
 from pathlib import Path
+from typing import Any
 
 # Add parent directory to path for imports
 sys.path.insert(0, str(Path(__file__).parent))
@@ -85,7 +86,7 @@ class BrowserRAG(BaseRAGExample):
 
         return profiles
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load browser history and convert to text chunks."""
         # Determine Chrome profiles
         if args.chrome_profile and not args.auto_find_profiles:

apps/chatgpt_rag.py

@@ -5,6 +5,7 @@ Supports ChatGPT export data from chat.html files.
 
 import sys
 from pathlib import Path
+from typing import Any
 
 # Add parent directory to path for imports
 sys.path.insert(0, str(Path(__file__).parent))
@@ -80,7 +81,7 @@ class ChatGPTRAG(BaseRAGExample):
 
         return export_files
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load ChatGPT export data and convert to text chunks."""
         export_path = Path(args.export_path)
 

apps/claude_rag.py

@@ -5,6 +5,7 @@ Supports Claude export data from JSON files.
 
 import sys
 from pathlib import Path
+from typing import Any
 
 # Add parent directory to path for imports
 sys.path.insert(0, str(Path(__file__).parent))
@@ -80,7 +81,7 @@ class ClaudeRAG(BaseRAGExample):
 
         return export_files
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load Claude export data and convert to text chunks."""
         export_path = Path(args.export_path)
 

apps/code_rag.py

@@ -6,6 +6,7 @@ optimized chunking parameters.
 
 import sys
 from pathlib import Path
+from typing import Any
 
 # Add parent directory to path for imports
 sys.path.insert(0, str(Path(__file__).parent))
@@ -77,7 +78,7 @@ class CodeRAG(BaseRAGExample):
             help="Try to preserve import statements in chunks (default: True)",
         )
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load code files and convert to AST-aware chunks."""
         print(f"🔍 Scanning code repository: {args.repo_dir}")
         print(f"📁 Including extensions: {args.include_extensions}")
@@ -88,14 +89,6 @@ class CodeRAG(BaseRAGExample):
         if not repo_path.exists():
             raise ValueError(f"Repository directory not found: {args.repo_dir}")
 
-        # Load code files with filtering
-        reader_kwargs = {
-            "recursive": True,
-            "encoding": "utf-8",
-            "required_exts": args.include_extensions,
-            "exclude_hidden": True,
-        }
-
         # Create exclusion filter
         def file_filter(file_path: str) -> bool:
             """Filter out unwanted files and directories."""
@@ -120,8 +113,11 @@ class CodeRAG(BaseRAGExample):
             # Load documents with file filtering
             documents = SimpleDirectoryReader(
                 args.repo_dir,
-                file_extractor=None,  # Use default extractors
+                file_extractor=None,
-                **reader_kwargs,
+                recursive=True,
+                encoding="utf-8",
+                required_exts=args.include_extensions,
+                exclude_hidden=True,
             ).load_data(show_progress=True)
 
             # Apply custom filtering

apps/colqwen_rag.py

@@ -0,0 +1,364 @@
+#!/usr/bin/env python3
+"""
+ColQwen RAG - Easy-to-use multimodal PDF retrieval with ColQwen2/ColPali
+
+Usage:
+    python -m apps.colqwen_rag build --pdfs ./my_pdfs/ --index my_index
+    python -m apps.colqwen_rag search my_index "How does attention work?"
+    python -m apps.colqwen_rag ask my_index --interactive
+"""
+
+import argparse
+import os
+import sys
+from pathlib import Path
+from typing import Optional, cast
+
+# Add LEANN packages to path
+_repo_root = Path(__file__).resolve().parents[1]
+_leann_core_src = _repo_root / "packages" / "leann-core" / "src"
+_leann_hnsw_pkg = _repo_root / "packages" / "leann-backend-hnsw"
+if str(_leann_core_src) not in sys.path:
+    sys.path.append(str(_leann_core_src))
+if str(_leann_hnsw_pkg) not in sys.path:
+    sys.path.append(str(_leann_hnsw_pkg))
+
+import torch  # noqa: E402
+from colpali_engine import ColPali, ColPaliProcessor, ColQwen2, ColQwen2Processor  # noqa: E402
+from colpali_engine.utils.torch_utils import ListDataset  # noqa: E402
+from pdf2image import convert_from_path  # noqa: E402
+from PIL import Image  # noqa: E402
+from torch.utils.data import DataLoader  # noqa: E402
+from tqdm import tqdm  # noqa: E402
+
+# Import the existing multi-vector implementation
+sys.path.append(str(_repo_root / "apps" / "multimodal" / "vision-based-pdf-multi-vector"))
+from leann_multi_vector import LeannMultiVector  # noqa: E402
+
+
+class ColQwenRAG:
+    """Easy-to-use ColQwen RAG system for multimodal PDF retrieval."""
+
+    def __init__(self, model_type: str = "colpali"):
+        """
+        Initialize ColQwen RAG system.
+
+        Args:
+            model_type: "colqwen2" or "colpali"
+        """
+        self.model_type = model_type
+        self.device = self._get_device()
+        # Use float32 on MPS to avoid memory issues, float16 on CUDA, bfloat16 on CPU
+        if self.device.type == "mps":
+            self.dtype = torch.float32
+        elif self.device.type == "cuda":
+            self.dtype = torch.float16
+        else:
+            self.dtype = torch.bfloat16
+
+        print(f"🚀 Initializing {model_type.upper()} on {self.device} with {self.dtype}")
+
+        # Load model and processor with MPS-optimized settings
+        try:
+            if model_type == "colqwen2":
+                self.model_name = "vidore/colqwen2-v1.0"
+                if self.device.type == "mps":
+                    # For MPS, load on CPU first then move to avoid memory allocation issues
+                    self.model = ColQwen2.from_pretrained(
+                        self.model_name,
+                        torch_dtype=self.dtype,
+                        device_map="cpu",
+                        low_cpu_mem_usage=True,
+                    ).eval()
+                    self.model = self.model.to(self.device)
+                else:
+                    self.model = ColQwen2.from_pretrained(
+                        self.model_name,
+                        torch_dtype=self.dtype,
+                        device_map=self.device,
+                        low_cpu_mem_usage=True,
+                    ).eval()
+                self.processor = ColQwen2Processor.from_pretrained(self.model_name)
+            else:  # colpali
+                self.model_name = "vidore/colpali-v1.2"
+                if self.device.type == "mps":
+                    # For MPS, load on CPU first then move to avoid memory allocation issues
+                    self.model = ColPali.from_pretrained(
+                        self.model_name,
+                        torch_dtype=self.dtype,
+                        device_map="cpu",
+                        low_cpu_mem_usage=True,
+                    ).eval()
+                    self.model = self.model.to(self.device)
+                else:
+                    self.model = ColPali.from_pretrained(
+                        self.model_name,
+                        torch_dtype=self.dtype,
+                        device_map=self.device,
+                        low_cpu_mem_usage=True,
+                    ).eval()
+                self.processor = ColPaliProcessor.from_pretrained(self.model_name)
+        except Exception as e:
+            if "memory" in str(e).lower() or "offload" in str(e).lower():
+                print(f"⚠️  Memory constraint on {self.device}, using CPU with optimizations...")
+                self.device = torch.device("cpu")
+                self.dtype = torch.float32
+
+                if model_type == "colqwen2":
+                    self.model = ColQwen2.from_pretrained(
+                        self.model_name,
+                        torch_dtype=self.dtype,
+                        device_map="cpu",
+                        low_cpu_mem_usage=True,
+                    ).eval()
+                else:
+                    self.model = ColPali.from_pretrained(
+                        self.model_name,
+                        torch_dtype=self.dtype,
+                        device_map="cpu",
+                        low_cpu_mem_usage=True,
+                    ).eval()
+            else:
+                raise
+
+    def _get_device(self):
+        """Auto-select best available device."""
+        if torch.cuda.is_available():
+            return torch.device("cuda")
+        elif hasattr(torch.backends, "mps") and torch.backends.mps.is_available():
+            return torch.device("mps")
+        else:
+            return torch.device("cpu")
+
+    def build_index(self, pdf_paths: list[str], index_name: str, pages_dir: Optional[str] = None):
+        """
+        Build multimodal index from PDF files.
+
+        Args:
+            pdf_paths: List of PDF file paths
+            index_name: Name for the index
+            pages_dir: Directory to save page images (optional)
+        """
+        print(f"Building index '{index_name}' from {len(pdf_paths)} PDFs...")
+
+        # Convert PDFs to images
+        all_images = []
+        all_metadata = []
+
+        if pages_dir:
+            os.makedirs(pages_dir, exist_ok=True)
+
+        for pdf_path in tqdm(pdf_paths, desc="Converting PDFs"):
+            try:
+                images = convert_from_path(pdf_path, dpi=150)
+                pdf_name = Path(pdf_path).stem
+
+                for i, image in enumerate(images):
+                    # Save image if pages_dir specified
+                    if pages_dir:
+                        image_path = Path(pages_dir) / f"{pdf_name}_page_{i + 1}.png"
+                        image.save(image_path)
+
+                    all_images.append(image)
+                    all_metadata.append(
+                        {
+                            "pdf_path": pdf_path,
+                            "pdf_name": pdf_name,
+                            "page_number": i + 1,
+                            "image_path": str(image_path) if pages_dir else None,
+                        }
+                    )
+
+            except Exception as e:
+                print(f"❌ Error processing {pdf_path}: {e}")
+                continue
+
+        print(f"📄 Converted {len(all_images)} pages from {len(pdf_paths)} PDFs")
+        print(f"All metadata: {all_metadata}")
+
+        # Generate embeddings
+        print("🧠 Generating embeddings...")
+        embeddings = self._embed_images(all_images)
+
+        # Build LEANN index
+        print("🔍 Building LEANN index...")
+        leann_mv = LeannMultiVector(
+            index_path=index_name,
+            dim=embeddings.shape[-1],
+            embedding_model_name=self.model_type,
+        )
+
+        # Create collection and insert data
+        leann_mv.create_collection()
+        for i, (embedding, metadata) in enumerate(zip(embeddings, all_metadata)):
+            data = {
+                "doc_id": i,
+                "filepath": metadata.get("image_path", ""),
+                "colbert_vecs": embedding.numpy(),  # Convert tensor to numpy
+            }
+            leann_mv.insert(data)
+
+        # Build the index
+        leann_mv.create_index()
+        print(f"✅ Index '{index_name}' built successfully!")
+
+        return leann_mv
+
+    def search(self, index_name: str, query: str, top_k: int = 5):
+        """
+        Search the index with a text query.
+
+        Args:
+            index_name: Name of the index to search
+            query: Text query
+            top_k: Number of results to return
+        """
+        print(f"🔍 Searching '{index_name}' for: '{query}'")
+
+        # Load index
+        leann_mv = LeannMultiVector(
+            index_path=index_name,
+            dim=128,  # Will be updated when loading
+            embedding_model_name=self.model_type,
+        )
+
+        # Generate query embedding
+        query_embedding = self._embed_query(query)
+
+        # Search (returns list of (score, doc_id) tuples)
+        search_results = leann_mv.search(query_embedding.numpy(), topk=top_k)
+
+        # Display results
+        print(f"\n📋 Top {len(search_results)} results:")
+        for i, (score, doc_id) in enumerate(search_results, 1):
+            # Get metadata for this doc_id (we need to load the metadata)
+            print(f"{i}. Score: {score:.3f} | Doc ID: {doc_id}")
+
+        return search_results
+
+    def ask(self, index_name: str, interactive: bool = False):
+        """
+        Interactive Q&A with the indexed documents.
+
+        Args:
+            index_name: Name of the index to query
+            interactive: Whether to run in interactive mode
+        """
+        print(f"💬 ColQwen Chat with '{index_name}'")
+
+        if interactive:
+            print("Type 'quit' to exit, 'help' for commands")
+            while True:
+                try:
+                    query = input("\n🤔 Your question: ").strip()
+                    if query.lower() in ["quit", "exit", "q"]:
+                        break
+                    elif query.lower() == "help":
+                        print("Commands: quit/exit/q (exit), help (this message)")
+                        continue
+                    elif not query:
+                        continue
+
+                    self.search(index_name, query, top_k=3)
+
+                    # TODO: Add answer generation with Qwen-VL
+                    print("\n💡 For detailed answers, we can integrate Qwen-VL here!")
+
+                except KeyboardInterrupt:
+                    print("\n👋 Goodbye!")
+                    break
+        else:
+            query = input("🤔 Your question: ").strip()
+            if query:
+                self.search(index_name, query)
+
+    def _embed_images(self, images: list[Image.Image]) -> torch.Tensor:
+        """Generate embeddings for a list of images."""
+        dataset = ListDataset(images)
+        dataloader = DataLoader(dataset, batch_size=1, shuffle=False, collate_fn=lambda x: x)
+
+        embeddings = []
+        with torch.no_grad():
+            for batch in tqdm(dataloader, desc="Embedding images"):
+                batch_images = cast(list, batch)
+                batch_inputs = self.processor.process_images(batch_images).to(self.device)
+                batch_embeddings = self.model(**batch_inputs)
+                embeddings.append(batch_embeddings.cpu())
+
+        return torch.cat(embeddings, dim=0)
+
+    def _embed_query(self, query: str) -> torch.Tensor:
+        """Generate embedding for a text query."""
+        with torch.no_grad():
+            query_inputs = self.processor.process_queries([query]).to(self.device)
+            query_embedding = self.model(**query_inputs)
+            return query_embedding.cpu()
+
+
+def main():
+    parser = argparse.ArgumentParser(description="ColQwen RAG - Easy multimodal PDF retrieval")
+    subparsers = parser.add_subparsers(dest="command", help="Available commands")
+
+    # Build command
+    build_parser = subparsers.add_parser("build", help="Build index from PDFs")
+    build_parser.add_argument("--pdfs", required=True, help="Directory containing PDF files")
+    build_parser.add_argument("--index", required=True, help="Index name")
+    build_parser.add_argument(
+        "--model", choices=["colqwen2", "colpali"], default="colqwen2", help="Model to use"
+    )
+    build_parser.add_argument("--pages-dir", help="Directory to save page images")
+
+    # Search command
+    search_parser = subparsers.add_parser("search", help="Search the index")
+    search_parser.add_argument("index", help="Index name")
+    search_parser.add_argument("query", help="Search query")
+    search_parser.add_argument("--top-k", type=int, default=5, help="Number of results")
+    search_parser.add_argument(
+        "--model", choices=["colqwen2", "colpali"], default="colqwen2", help="Model to use"
+    )
+
+    # Ask command
+    ask_parser = subparsers.add_parser("ask", help="Interactive Q&A")
+    ask_parser.add_argument("index", help="Index name")
+    ask_parser.add_argument("--interactive", action="store_true", help="Interactive mode")
+    ask_parser.add_argument(
+        "--model", choices=["colqwen2", "colpali"], default="colqwen2", help="Model to use"
+    )
+
+    args = parser.parse_args()
+
+    if not args.command:
+        parser.print_help()
+        return
+
+    # Initialize ColQwen RAG
+    if args.command == "build":
+        colqwen = ColQwenRAG(args.model)
+
+        # Get PDF files
+        pdf_dir = Path(args.pdfs)
+        if pdf_dir.is_file() and pdf_dir.suffix.lower() == ".pdf":
+            pdf_paths = [str(pdf_dir)]
+        elif pdf_dir.is_dir():
+            pdf_paths = [str(p) for p in pdf_dir.glob("*.pdf")]
+        else:
+            print(f"❌ Invalid PDF path: {args.pdfs}")
+            return
+
+        if not pdf_paths:
+            print(f"❌ No PDF files found in {args.pdfs}")
+            return
+
+        colqwen.build_index(pdf_paths, args.index, args.pages_dir)
+
+    elif args.command == "search":
+        colqwen = ColQwenRAG(args.model)
+        colqwen.search(args.index, args.query, args.top_k)
+
+    elif args.command == "ask":
+        colqwen = ColQwenRAG(args.model)
+        colqwen.ask(args.index, args.interactive)
+
+
+if __name__ == "__main__":
+    main()

apps/document_rag.py

@@ -5,6 +5,7 @@ Supports PDF, TXT, MD, and other document formats.
 
 import sys
 from pathlib import Path
+from typing import Any
 
 # Add parent directory to path for imports
 sys.path.insert(0, str(Path(__file__).parent))
@@ -51,7 +52,7 @@ class DocumentRAG(BaseRAGExample):
             help="Enable AST-aware chunking for code files in the data directory",
         )
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load documents and convert to text chunks."""
         print(f"Loading documents from: {args.data_dir}")
         if args.file_types:
@@ -65,16 +66,12 @@ class DocumentRAG(BaseRAGExample):
             raise ValueError(f"Data directory not found: {args.data_dir}")
 
         # Load documents
-        reader_kwargs = {
+        documents = SimpleDirectoryReader(
-            "recursive": True,
+            args.data_dir,
-            "encoding": "utf-8",
+            recursive=True,
-        }
+            encoding="utf-8",
-        if args.file_types:
+            required_exts=args.file_types if args.file_types else None,
-            reader_kwargs["required_exts"] = args.file_types
+        ).load_data(show_progress=True)
-
-        documents = SimpleDirectoryReader(args.data_dir, **reader_kwargs).load_data(
-            show_progress=True
-        )
 
         if not documents:
             print(f"No documents found in {args.data_dir} with extensions {args.file_types}")

apps/email_data/email.py

@@ -127,11 +127,12 @@ class EmlxMboxReader(MboxReader):
 
     def load_data(
         self,
-        directory: Path,
+        file: Path,  # Note: for EmlxMboxReader, this is actually a directory
         extra_info: dict | None = None,
         fs: AbstractFileSystem | None = None,
     ) -> list[Document]:
         """Parse .emlx files from directory into strings using MboxReader logic."""
+        directory = file  # Rename for clarity - this is a directory of .emlx files
         import os
         import tempfile
 

apps/email_rag.py

@@ -5,6 +5,7 @@ Supports Apple Mail on macOS.
 
 import sys
 from pathlib import Path
+from typing import Any
 
 # Add parent directory to path for imports
 sys.path.insert(0, str(Path(__file__).parent))
@@ -64,7 +65,7 @@ class EmailRAG(BaseRAGExample):
 
         return messages_dirs
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load emails and convert to text chunks."""
         # Determine mail directories
         if args.mail_path:

apps/history_data/wechat_history.py

@@ -86,7 +86,7 @@ class WeChatHistoryReader(BaseReader):
                 text=True,
                 timeout=5,
             )
-            return result.returncode == 0 and result.stdout.strip()
+            return result.returncode == 0 and bool(result.stdout.strip())
         except Exception:
             return False
 
@@ -314,7 +314,9 @@ class WeChatHistoryReader(BaseReader):
 
         return concatenated_groups
 
-    def _create_concatenated_content(self, message_group: dict, contact_name: str) -> str:
+    def _create_concatenated_content(
+        self, message_group: dict, contact_name: str
+    ) -> tuple[str, str]:
         """
         Create concatenated content from a group of messages.
 

apps/image_rag.py

@@ -0,0 +1,219 @@
+#!/usr/bin/env python3
+"""
+CLIP Image RAG Application
+
+This application enables RAG (Retrieval-Augmented Generation) on images using CLIP embeddings.
+You can index a directory of images and search them using text queries.
+
+Usage:
+    python -m apps.image_rag --image-dir ./my_images/ --query "a sunset over mountains"
+    python -m apps.image_rag --image-dir ./my_images/ --interactive
+"""
+
+import argparse
+import pickle
+import tempfile
+from pathlib import Path
+from typing import Any
+
+import numpy as np
+from PIL import Image
+from sentence_transformers import SentenceTransformer
+from tqdm import tqdm
+
+from apps.base_rag_example import BaseRAGExample
+
+
+class ImageRAG(BaseRAGExample):
+    """
+    RAG application for images using CLIP embeddings.
+
+    This class provides a complete RAG pipeline for image data, including
+    CLIP embedding generation, indexing, and text-based image search.
+    """
+
+    def __init__(self):
+        super().__init__(
+            name="Image RAG",
+            description="RAG application for images using CLIP embeddings",
+            default_index_name="image_index",
+        )
+        # Override default embedding model to use CLIP
+        self.embedding_model_default = "clip-ViT-L-14"
+        self.embedding_mode_default = "sentence-transformers"
+        self._image_data: list[dict] = []
+
+    def _add_specific_arguments(self, parser: argparse.ArgumentParser):
+        """Add image-specific arguments."""
+        image_group = parser.add_argument_group("Image Parameters")
+        image_group.add_argument(
+            "--image-dir",
+            type=str,
+            required=True,
+            help="Directory containing images to index",
+        )
+        image_group.add_argument(
+            "--image-extensions",
+            type=str,
+            nargs="+",
+            default=[".jpg", ".jpeg", ".png", ".gif", ".bmp", ".webp"],
+            help="Image file extensions to process (default: .jpg .jpeg .png .gif .bmp .webp)",
+        )
+        image_group.add_argument(
+            "--batch-size",
+            type=int,
+            default=32,
+            help="Batch size for CLIP embedding generation (default: 32)",
+        )
+
+    async def load_data(self, args) -> list[dict[str, Any]]:
+        """Load images, generate CLIP embeddings, and return text descriptions."""
+        self._image_data = self._load_images_and_embeddings(args)
+        return [entry["text"] for entry in self._image_data]
+
+    def _load_images_and_embeddings(self, args) -> list[dict]:
+        """Helper to process images and produce embeddings/metadata."""
+        image_dir = Path(args.image_dir)
+        if not image_dir.exists():
+            raise ValueError(f"Image directory does not exist: {image_dir}")
+
+        print(f"📸 Loading images from {image_dir}...")
+
+        # Find all image files
+        image_files = []
+        for ext in args.image_extensions:
+            image_files.extend(image_dir.rglob(f"*{ext}"))
+            image_files.extend(image_dir.rglob(f"*{ext.upper()}"))
+
+        if not image_files:
+            raise ValueError(
+                f"No images found in {image_dir} with extensions {args.image_extensions}"
+            )
+
+        print(f"✅ Found {len(image_files)} images")
+
+        # Limit if max_items is set
+        if args.max_items > 0:
+            image_files = image_files[: args.max_items]
+            print(f"📊 Processing {len(image_files)} images (limited by --max-items)")
+
+        # Load CLIP model
+        print("🔍 Loading CLIP model...")
+        model = SentenceTransformer(self.embedding_model_default)
+
+        # Process images and generate embeddings
+        print("🖼️  Processing images and generating embeddings...")
+        image_data = []
+        batch_images = []
+        batch_paths = []
+
+        for image_path in tqdm(image_files, desc="Processing images"):
+            try:
+                image = Image.open(image_path).convert("RGB")
+                batch_images.append(image)
+                batch_paths.append(image_path)
+
+                # Process in batches
+                if len(batch_images) >= args.batch_size:
+                    embeddings = model.encode(
+                        batch_images,
+                        convert_to_numpy=True,
+                        normalize_embeddings=True,
+                        batch_size=args.batch_size,
+                        show_progress_bar=False,
+                    )
+
+                    for img_path, embedding in zip(batch_paths, embeddings):
+                        image_data.append(
+                            {
+                                "text": f"Image: {img_path.name}\nPath: {img_path}",
+                                "metadata": {
+                                    "image_path": str(img_path),
+                                    "image_name": img_path.name,
+                                    "image_dir": str(image_dir),
+                                },
+                                "embedding": embedding.astype(np.float32),
+                            }
+                        )
+
+                    batch_images = []
+                    batch_paths = []
+
+            except Exception as e:
+                print(f"⚠️  Failed to process {image_path}: {e}")
+                continue
+
+        # Process remaining images
+        if batch_images:
+            embeddings = model.encode(
+                batch_images,
+                convert_to_numpy=True,
+                normalize_embeddings=True,
+                batch_size=len(batch_images),
+                show_progress_bar=False,
+            )
+
+            for img_path, embedding in zip(batch_paths, embeddings):
+                image_data.append(
+                    {
+                        "text": f"Image: {img_path.name}\nPath: {img_path}",
+                        "metadata": {
+                            "image_path": str(img_path),
+                            "image_name": img_path.name,
+                            "image_dir": str(image_dir),
+                        },
+                        "embedding": embedding.astype(np.float32),
+                    }
+                )
+
+        print(f"✅ Processed {len(image_data)} images")
+        return image_data
+
+    async def build_index(self, args, texts: list[dict[str, Any]]) -> str:
+        """Build index using pre-computed CLIP embeddings."""
+        from leann.api import LeannBuilder
+
+        if not self._image_data or len(self._image_data) != len(texts):
+            raise RuntimeError("No image data found. Make sure load_data() ran successfully.")
+
+        print("🔨 Building LEANN index with CLIP embeddings...")
+        builder = LeannBuilder(
+            backend_name=args.backend_name,
+            embedding_model=self.embedding_model_default,
+            embedding_mode=self.embedding_mode_default,
+            is_recompute=False,
+            distance_metric="cosine",
+            graph_degree=args.graph_degree,
+            build_complexity=args.build_complexity,
+            is_compact=not args.no_compact,
+        )
+
+        for text, data in zip(texts, self._image_data):
+            builder.add_text(text=text, metadata=data["metadata"])
+
+        ids = [str(i) for i in range(len(self._image_data))]
+        embeddings = np.array([data["embedding"] for data in self._image_data], dtype=np.float32)
+
+        with tempfile.NamedTemporaryFile(mode="wb", suffix=".pkl", delete=False) as f:
+            pickle.dump((ids, embeddings), f)
+            pkl_path = f.name
+
+        try:
+            index_path = str(Path(args.index_dir) / f"{self.default_index_name}.leann")
+            builder.build_index_from_embeddings(index_path, pkl_path)
+            print(f"✅ Index built successfully at {index_path}")
+            return index_path
+        finally:
+            Path(pkl_path).unlink()
+
+
+def main():
+    """Main entry point for the image RAG application."""
+    import asyncio
+
+    app = ImageRAG()
+    asyncio.run(app.run())
+
+
+if __name__ == "__main__":
+    main()

apps/imessage_rag.py

@@ -6,6 +6,7 @@ This example demonstrates how to build a RAG system on your iMessage conversatio
 
 import asyncio
 from pathlib import Path
+from typing import Any
 
 from leann.chunking_utils import create_text_chunks
 
@@ -56,7 +57,7 @@ class IMessageRAG(BaseRAGExample):
             help="Overlap between text chunks (default: 200)",
         )
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load iMessage history and convert to text chunks."""
         print("Loading iMessage conversation history...")
 

apps/multimodal/vision-based-pdf-multi-vector/multi-vector-leann-paper-example.py

@@ -18,10 +18,11 @@ _repo_root = Path(__file__).resolve().parents[3]
 _leann_core_src = _repo_root / "packages" / "leann-core" / "src"
 _leann_hnsw_pkg = _repo_root / "packages" / "leann-backend-hnsw"
 if str(_leann_core_src) not in sys.path:
-    sys.path.append(str(_leann_core_src))
+    sys.path.insert(0, str(_leann_core_src))
 if str(_leann_hnsw_pkg) not in sys.path:
-    sys.path.append(str(_leann_hnsw_pkg))
+    sys.path.insert(0, str(_leann_hnsw_pkg))
 
+from leann_multi_vector import LeannMultiVector
 
 import torch
 from colpali_engine.models import ColPali
@@ -93,9 +94,9 @@ for batch_doc in tqdm(dataloader):
 print(ds[0].shape)
 
 # %%
-# Build HNSW index via LeannRetriever primitives and run search
+# Build HNSW index via LeannMultiVector primitives and run search
 index_path = "./indexes/colpali.leann"
-retriever = LeannRetriever(index_path=index_path, dim=int(ds[0].shape[-1]))
+retriever = LeannMultiVector(index_path=index_path, dim=int(ds[0].shape[-1]))
 retriever.create_collection()
 filepaths = [os.path.join("./pages", name) for name in page_filenames]
 for i in range(len(filepaths)):

apps/multimodal/vision-based-pdf-multi-vector/multi-vector-leann-similarity-map.py

@@ -5,7 +5,7 @@ import argparse
 import faulthandler
 import os
 import time
-from typing import Any, Optional
+from typing import Any, Optional, cast
 
 import numpy as np
 from PIL import Image
@@ -223,7 +223,7 @@ if need_to_build_index:
         # Use filenames as identifiers instead of full paths for cleaner metadata
         filepaths = [os.path.basename(fp) for fp in filepaths]
     elif USE_HF_DATASET:
-        from datasets import load_dataset, concatenate_datasets, DatasetDict
+        from datasets import Dataset, DatasetDict, concatenate_datasets, load_dataset
 
         # Determine which datasets to load
         if DATASET_NAMES is not None:
@@ -281,12 +281,12 @@ if need_to_build_index:
                 splits_to_load = DATASET_SPLITS
 
             # Load and concatenate multiple splits for this dataset
-            datasets_to_concat = []
+            datasets_to_concat: list[Dataset] = []
             for split in splits_to_load:
                 if split not in dataset_dict:
                     print(f"  Warning: Split '{split}' not found in dataset. Available splits: {list(dataset_dict.keys())}")
                     continue
-                split_dataset = dataset_dict[split]
+                split_dataset = cast(Dataset, dataset_dict[split])
                 print(f"  Loaded split '{split}': {len(split_dataset)} pages")
                 datasets_to_concat.append(split_dataset)
 

apps/multimodal/vision-based-pdf-multi-vector/vidore_v1_benchmark.py

@@ -25,9 +25,9 @@ Usage:
 import argparse
 import json
 import os
-from typing import Optional
+from typing import Any, Optional, cast
 
-from datasets import load_dataset
+from datasets import Dataset, load_dataset
 from leann_multi_vector import (
     ViDoReBenchmarkEvaluator,
     _ensure_repo_paths_importable,
@@ -151,40 +151,43 @@ def load_vidore_v1_data(
     """
     print(f"Loading dataset: {dataset_path} (split={split})")
 
-    # Load queries
+    # Load queries - cast to Dataset since we know split returns Dataset not DatasetDict
-    query_ds = load_dataset(dataset_path, "queries", split=split, revision=revision)
+    query_ds = cast(Dataset, load_dataset(dataset_path, "queries", split=split, revision=revision))
 
-    queries = {}
+    queries: dict[str, str] = {}
     for row in query_ds:
-        query_id = f"query-{split}-{row['query-id']}"
+        row_dict = cast(dict[str, Any], row)
-        queries[query_id] = row["query"]
+        query_id = f"query-{split}-{row_dict['query-id']}"
+        queries[query_id] = row_dict["query"]
 
-    # Load corpus (images)
+    # Load corpus (images) - cast to Dataset
-    corpus_ds = load_dataset(dataset_path, "corpus", split=split, revision=revision)
+    corpus_ds = cast(Dataset, load_dataset(dataset_path, "corpus", split=split, revision=revision))
 
-    corpus = {}
+    corpus: dict[str, Any] = {}
     for row in corpus_ds:
-        corpus_id = f"corpus-{split}-{row['corpus-id']}"
+        row_dict = cast(dict[str, Any], row)
+        corpus_id = f"corpus-{split}-{row_dict['corpus-id']}"
         # Extract image from the dataset row
-        if "image" in row:
+        if "image" in row_dict:
-            corpus[corpus_id] = row["image"]
+            corpus[corpus_id] = row_dict["image"]
-        elif "page_image" in row:
+        elif "page_image" in row_dict:
-            corpus[corpus_id] = row["page_image"]
+            corpus[corpus_id] = row_dict["page_image"]
         else:
             raise ValueError(
-                f"No image field found in corpus. Available fields: {list(row.keys())}"
+                f"No image field found in corpus. Available fields: {list(row_dict.keys())}"
             )
 
-    # Load qrels (relevance judgments)
+    # Load qrels (relevance judgments) - cast to Dataset
-    qrels_ds = load_dataset(dataset_path, "qrels", split=split, revision=revision)
+    qrels_ds = cast(Dataset, load_dataset(dataset_path, "qrels", split=split, revision=revision))
 
-    qrels = {}
+    qrels: dict[str, dict[str, int]] = {}
     for row in qrels_ds:
-        query_id = f"query-{split}-{row['query-id']}"
+        row_dict = cast(dict[str, Any], row)
-        corpus_id = f"corpus-{split}-{row['corpus-id']}"
+        query_id = f"query-{split}-{row_dict['query-id']}"
+        corpus_id = f"corpus-{split}-{row_dict['corpus-id']}"
         if query_id not in qrels:
             qrels[query_id] = {}
-        qrels[query_id][corpus_id] = int(row["score"])
+        qrels[query_id][corpus_id] = int(row_dict["score"])
 
     print(
         f"Loaded {len(queries)} queries, {len(corpus)} corpus items, {len(qrels)} query-relevance mappings"
@@ -234,8 +237,8 @@ def evaluate_task(
         raise ValueError(f"Unknown task: {task_name}. Available: {list(VIDORE_V1_TASKS.keys())}")
 
     task_config = VIDORE_V1_TASKS[task_name]
-    dataset_path = task_config["dataset_path"]
+    dataset_path = str(task_config["dataset_path"])
-    revision = task_config["revision"]
+    revision = str(task_config["revision"])
 
     # Load data
     corpus, queries, qrels = load_vidore_v1_data(
@@ -286,7 +289,7 @@ def evaluate_task(
     )
 
     # Search queries
-    task_prompt = task_config.get("prompt")
+    task_prompt = cast(Optional[dict[str, str]], task_config.get("prompt"))
     results = evaluator.search_queries(
         queries=queries,
         corpus_ids=corpus_ids_ordered,

apps/multimodal/vision-based-pdf-multi-vector/vidore_v2_benchmark.py

@@ -25,9 +25,9 @@ Usage:
 import argparse
 import json
 import os
-from typing import Optional
+from typing import Any, Optional, cast
 
-from datasets import load_dataset
+from datasets import Dataset, load_dataset
 from leann_multi_vector import (
     ViDoReBenchmarkEvaluator,
     _ensure_repo_paths_importable,
@@ -91,8 +91,8 @@ def load_vidore_v2_data(
     """
     print(f"Loading dataset: {dataset_path} (split={split}, language={language})")
 
-    # Load queries
+    # Load queries - cast to Dataset since we know split returns Dataset not DatasetDict
-    query_ds = load_dataset(dataset_path, "queries", split=split, revision=revision)
+    query_ds = cast(Dataset, load_dataset(dataset_path, "queries", split=split, revision=revision))
 
     # Check if dataset has language field before filtering
     has_language_field = len(query_ds) > 0 and "language" in query_ds.column_names
@@ -112,8 +112,9 @@ def load_vidore_v2_data(
             if len(query_ds_filtered) == 0:
                 # Try to get a sample to see actual language values
                 try:
-                    sample_ds = load_dataset(
+                    sample_ds = cast(
-                        dataset_path, "queries", split=split, revision=revision
+                        Dataset,
+                        load_dataset(dataset_path, "queries", split=split, revision=revision),
                     )
                     if len(sample_ds) > 0 and "language" in sample_ds.column_names:
                         sample_langs = set(sample_ds["language"])
@@ -126,37 +127,40 @@ def load_vidore_v2_data(
                 )
         query_ds = query_ds_filtered
 
-    queries = {}
+    queries: dict[str, str] = {}
     for row in query_ds:
-        query_id = f"query-{split}-{row['query-id']}"
+        row_dict = cast(dict[str, Any], row)
-        queries[query_id] = row["query"]
+        query_id = f"query-{split}-{row_dict['query-id']}"
+        queries[query_id] = row_dict["query"]
 
-    # Load corpus (images)
+    # Load corpus (images) - cast to Dataset
-    corpus_ds = load_dataset(dataset_path, "corpus", split=split, revision=revision)
+    corpus_ds = cast(Dataset, load_dataset(dataset_path, "corpus", split=split, revision=revision))
 
-    corpus = {}
+    corpus: dict[str, Any] = {}
     for row in corpus_ds:
-        corpus_id = f"corpus-{split}-{row['corpus-id']}"
+        row_dict = cast(dict[str, Any], row)
+        corpus_id = f"corpus-{split}-{row_dict['corpus-id']}"
         # Extract image from the dataset row
-        if "image" in row:
+        if "image" in row_dict:
-            corpus[corpus_id] = row["image"]
+            corpus[corpus_id] = row_dict["image"]
-        elif "page_image" in row:
+        elif "page_image" in row_dict:
-            corpus[corpus_id] = row["page_image"]
+            corpus[corpus_id] = row_dict["page_image"]
         else:
             raise ValueError(
-                f"No image field found in corpus. Available fields: {list(row.keys())}"
+                f"No image field found in corpus. Available fields: {list(row_dict.keys())}"
             )
 
-    # Load qrels (relevance judgments)
+    # Load qrels (relevance judgments) - cast to Dataset
-    qrels_ds = load_dataset(dataset_path, "qrels", split=split, revision=revision)
+    qrels_ds = cast(Dataset, load_dataset(dataset_path, "qrels", split=split, revision=revision))
 
-    qrels = {}
+    qrels: dict[str, dict[str, int]] = {}
     for row in qrels_ds:
-        query_id = f"query-{split}-{row['query-id']}"
+        row_dict = cast(dict[str, Any], row)
-        corpus_id = f"corpus-{split}-{row['corpus-id']}"
+        query_id = f"query-{split}-{row_dict['query-id']}"
+        corpus_id = f"corpus-{split}-{row_dict['corpus-id']}"
         if query_id not in qrels:
             qrels[query_id] = {}
-        qrels[query_id][corpus_id] = int(row["score"])
+        qrels[query_id][corpus_id] = int(row_dict["score"])
 
     print(
         f"Loaded {len(queries)} queries, {len(corpus)} corpus items, {len(qrels)} query-relevance mappings"
@@ -204,13 +208,13 @@ def evaluate_task(
         raise ValueError(f"Unknown task: {task_name}. Available: {list(VIDORE_V2_TASKS.keys())}")
 
     task_config = VIDORE_V2_TASKS[task_name]
-    dataset_path = task_config["dataset_path"]
+    dataset_path = str(task_config["dataset_path"])
-    revision = task_config["revision"]
+    revision = str(task_config["revision"])
 
     # Determine language
     if language is None:
         # Use first language if multiple available
-        languages = task_config.get("languages")
+        languages = cast(Optional[list[str]], task_config.get("languages"))
         if languages is None:
             # Task doesn't support language filtering (e.g., Vidore2ESGReportsHLRetrieval)
             language = None
@@ -269,7 +273,7 @@ def evaluate_task(
     )
 
     # Search queries
-    task_prompt = task_config.get("prompt")
+    task_prompt = cast(Optional[dict[str, str]], task_config.get("prompt"))
     results = evaluator.search_queries(
         queries=queries,
         corpus_ids=corpus_ids_ordered,

apps/slack_data/slack_mcp_reader.py

@@ -177,7 +177,9 @@ class SlackMCPReader:
                     break
 
         # If we get here, all retries failed or it's not a retryable error
+        if last_exception is not None:
             raise last_exception
+        raise RuntimeError("Unexpected error: no exception captured during retry loop")
 
     async def fetch_slack_messages(
         self, channel: Optional[str] = None, limit: int = 100
@@ -267,7 +269,10 @@ class SlackMCPReader:
                     messages = json.loads(content["text"])
                 except json.JSONDecodeError:
                     # If not JSON, try to parse as CSV format (Slack MCP server format)
-                    messages = self._parse_csv_messages(content["text"], channel)
+                    text_content = content.get("text", "")
+                    messages = self._parse_csv_messages(
+                        text_content if text_content else "", channel or "unknown"
+                    )
             else:
                 messages = result["content"]
         else:

apps/slack_rag.py

@@ -11,6 +11,7 @@ Usage:
 
 import argparse
 import asyncio
+from typing import Any
 
 from apps.base_rag_example import BaseRAGExample
 from apps.slack_data.slack_mcp_reader import SlackMCPReader
@@ -139,7 +140,7 @@ class SlackMCPRAG(BaseRAGExample):
             print("4. Try running the MCP server command directly to test it")
             return False
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load Slack messages via MCP server."""
         print(f"Connecting to Slack MCP server: {args.mcp_server}")
 
@@ -188,7 +189,8 @@ class SlackMCPRAG(BaseRAGExample):
                 print(sample_text)
                 print("-" * 40)
 
-            return texts
+            # Convert strings to dict format expected by base class
+            return [{"text": text, "metadata": {"source": "slack"}} for text in texts]
 
         except Exception as e:
             print(f"Error loading Slack data: {e}")

apps/twitter_rag.py

@@ -11,6 +11,7 @@ Usage:
 
 import argparse
 import asyncio
+from typing import Any
 
 from apps.base_rag_example import BaseRAGExample
 from apps.twitter_data.twitter_mcp_reader import TwitterMCPReader
@@ -116,7 +117,7 @@ class TwitterMCPRAG(BaseRAGExample):
             print("5. Try running the MCP server command directly to test it")
             return False
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load Twitter bookmarks via MCP server."""
         print(f"Connecting to Twitter MCP server: {args.mcp_server}")
 
@@ -156,7 +157,8 @@ class TwitterMCPRAG(BaseRAGExample):
                 print(sample_text)
                 print("-" * 50)
 
-            return texts
+            # Convert strings to dict format expected by base class
+            return [{"text": text, "metadata": {"source": "twitter"}} for text in texts]
 
         except Exception as e:
             print(f"❌ Error loading Twitter bookmarks: {e}")

apps/wechat_rag.py

@@ -6,6 +6,7 @@ Supports WeChat chat history export and search.
 import subprocess
 import sys
 from pathlib import Path
+from typing import Any
 
 # Add parent directory to path for imports
 sys.path.insert(0, str(Path(__file__).parent))
@@ -91,7 +92,7 @@ class WeChatRAG(BaseRAGExample):
             print(f"Export error: {e}")
             return False
 
-    async def load_data(self, args) -> list[str]:
+    async def load_data(self, args) -> list[dict[str, Any]]:
         """Load WeChat history and convert to text chunks."""
         # Initialize WeChat reader with export capabilities
         reader = WeChatHistoryReader()

docs/COLQWEN_GUIDE.md

@@ -0,0 +1,200 @@
+# ColQwen Integration Guide
+
+Easy-to-use multimodal PDF retrieval with ColQwen2/ColPali models.
+
+## Quick Start
+
+> **🍎 Mac Users**: ColQwen is optimized for Apple Silicon with MPS acceleration for faster inference!
+
+### 1. Install Dependencies
+```bash
+uv pip install colpali_engine pdf2image pillow matplotlib qwen_vl_utils einops seaborn
+brew install poppler  # macOS only, for PDF processing
+```
+
+### 2. Basic Usage
+```bash
+# Build index from PDFs
+python -m apps.colqwen_rag build --pdfs ./my_papers/ --index research_papers
+
+# Search with text queries
+python -m apps.colqwen_rag search research_papers "How does attention mechanism work?"
+
+# Interactive Q&A
+python -m apps.colqwen_rag ask research_papers --interactive
+```
+
+## Commands
+
+### Build Index
+```bash
+python -m apps.colqwen_rag build \
+  --pdfs ./pdf_directory/ \
+  --index my_index \
+  --model colqwen2 \
+  --pages-dir ./page_images/  # Optional: save page images
+```
+
+**Options:**
+- `--pdfs`: Directory containing PDF files (or single PDF path)
+- `--index`: Name for the index (required)
+- `--model`: `colqwen2` (default) or `colpali`
+- `--pages-dir`: Directory to save page images (optional)
+
+### Search Index
+```bash
+python -m apps.colqwen_rag search my_index "your question here" --top-k 5
+```
+
+**Options:**
+- `--top-k`: Number of results to return (default: 5)
+- `--model`: Model used for search (should match build model)
+
+### Interactive Q&A
+```bash
+python -m apps.colqwen_rag ask my_index --interactive
+```
+
+**Commands in interactive mode:**
+- Type your questions naturally
+- `help`: Show available commands
+- `quit`/`exit`/`q`: Exit interactive mode
+
+## 🧪 Test & Reproduce Results
+
+Run the reproduction test for issue #119:
+```bash
+python test_colqwen_reproduction.py
+```
+
+This will:
+1. ✅ Check dependencies
+2. 📥 Download sample PDF (Attention Is All You Need paper)
+3. 🏗️ Build test index
+4. 🔍 Run sample queries
+5. 📊 Show how to generate similarity maps
+
+## 🎨 Advanced: Similarity Maps
+
+For visual similarity analysis, use the existing advanced script:
+```bash
+cd apps/multimodal/vision-based-pdf-multi-vector/
+python multi-vector-leann-similarity-map.py
+```
+
+Edit the script to customize:
+- `QUERY`: Your question
+- `MODEL`: "colqwen2" or "colpali"
+- `USE_HF_DATASET`: Use HuggingFace dataset or local PDFs
+- `SIMILARITY_MAP`: Generate heatmaps
+- `ANSWER`: Enable Qwen-VL answer generation
+
+## 🔧 How It Works
+
+### ColQwen2 vs ColPali
+- **ColQwen2** (`vidore/colqwen2-v1.0`): Latest vision-language model
+- **ColPali** (`vidore/colpali-v1.2`): Proven multimodal retriever
+
+### Architecture
+1. **PDF → Images**: Convert PDF pages to images (150 DPI)
+2. **Vision Encoding**: Process images with ColQwen2/ColPali
+3. **Multi-Vector Index**: Build LEANN HNSW index with multiple embeddings per page
+4. **Query Processing**: Encode text queries with same model
+5. **Similarity Search**: Find most relevant pages/regions
+6. **Visual Maps**: Generate attention heatmaps (optional)
+
+### Device Support
+- **CUDA**: Best performance with GPU acceleration
+- **MPS**: Apple Silicon Mac support
+- **CPU**: Fallback for any system (slower)
+
+Auto-detection: CUDA > MPS > CPU
+
+## 📊 Performance Tips
+
+### For Best Performance:
+```bash
+# Use ColQwen2 for latest features
+--model colqwen2
+
+# Save page images for reuse
+--pages-dir ./cached_pages/
+
+# Adjust batch size based on GPU memory
+# (automatically handled)
+```
+
+### For Large Document Sets:
+- Process PDFs in batches
+- Use SSD storage for index files
+- Consider using CUDA if available
+
+## 🔗 Related Resources
+
+- **Fast-PLAID**: https://github.com/lightonai/fast-plaid
+- **Pylate**: https://github.com/lightonai/pylate
+- **ColBERT**: https://github.com/stanford-futuredata/ColBERT
+- **ColPali Paper**: Vision-Language Models for Document Retrieval
+- **Issue #119**: https://github.com/yichuan-w/LEANN/issues/119
+
+## 🐛 Troubleshooting
+
+### PDF Conversion Issues (macOS)
+```bash
+# Install poppler
+brew install poppler
+which pdfinfo && pdfinfo -v
+```
+
+### Memory Issues
+- Reduce batch size (automatically handled)
+- Use CPU instead of GPU: `export CUDA_VISIBLE_DEVICES=""`
+- Process fewer PDFs at once
+
+### Model Download Issues
+- Ensure internet connection for first run
+- Models are cached after first download
+- Use HuggingFace mirrors if needed
+
+### Import Errors
+```bash
+# Ensure all dependencies installed
+uv pip install colpali_engine pdf2image pillow matplotlib qwen_vl_utils einops seaborn
+
+# Check PyTorch installation
+python -c "import torch; print(torch.__version__)"
+```
+
+## 💡 Examples
+
+### Research Paper Analysis
+```bash
+# Index your research papers
+python -m apps.colqwen_rag build --pdfs ~/Papers/AI/ --index ai_papers
+
+# Ask research questions
+python -m apps.colqwen_rag search ai_papers "What are the limitations of transformer models?"
+python -m apps.colqwen_rag search ai_papers "How does BERT compare to GPT?"
+```
+
+### Document Q&A
+```bash
+# Index business documents
+python -m apps.colqwen_rag build --pdfs ~/Documents/Reports/ --index reports
+
+# Interactive analysis
+python -m apps.colqwen_rag ask reports --interactive
+```
+
+### Visual Analysis
+```bash
+# Generate similarity maps for specific queries
+cd apps/multimodal/vision-based-pdf-multi-vector/
+# Edit multi-vector-leann-similarity-map.py with your query
+python multi-vector-leann-similarity-map.py
+# Check ./figures/ for generated heatmaps
+```
+
+---
+
+**🎯 This integration makes ColQwen as easy to use as other LEANN features while maintaining the full power of multimodal document understanding!**

packages/leann-core/pyproject.toml

@@ -7,7 +7,7 @@ name = "leann-core"
 version = "0.3.5"
 description = "Core API and plugin system for LEANN"
 readme = "README.md"
-requires-python = ">=3.9"
+requires-python = ">=3.10"
 license = { text = "MIT" }
 
 # All required dependencies included

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

@@ -239,11 +239,11 @@ def create_ast_chunks(
 
             chunks = chunk_builder.chunkify(code_content)
             for chunk in chunks:
-                chunk_text = None
+                chunk_text: str | None = None
-                astchunk_metadata = {}
+                astchunk_metadata: dict[str, Any] = {}
 
                 if hasattr(chunk, "text"):
-                    chunk_text = chunk.text
+                    chunk_text = str(chunk.text) if chunk.text else None
                 elif isinstance(chunk, str):
                     chunk_text = chunk
                 elif isinstance(chunk, dict):

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

@@ -19,7 +19,7 @@ from .settings import (
 )
 
 
-def extract_pdf_text_with_pymupdf(file_path: str) -> str:
+def extract_pdf_text_with_pymupdf(file_path: str) -> str | None:
     """Extract text from PDF using PyMuPDF for better quality."""
     try:
         import fitz  # PyMuPDF
@@ -35,7 +35,7 @@ def extract_pdf_text_with_pymupdf(file_path: str) -> str:
         return None
 
 
-def extract_pdf_text_with_pdfplumber(file_path: str) -> str:
+def extract_pdf_text_with_pdfplumber(file_path: str) -> str | None:
     """Extract text from PDF using pdfplumber for better quality."""
     try:
         import pdfplumber

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

@@ -451,7 +451,8 @@ def compute_embeddings_sentence_transformers(
             # TODO: Haven't tested this yet
             torch.set_num_threads(min(8, os.cpu_count() or 4))
             try:
-                torch.backends.mkldnn.enabled = True
+                # PyTorch's ContextProp type is complex; cast for type checker
+                torch.backends.mkldnn.enabled = True  # type: ignore[assignment]
             except AttributeError:
                 pass
 

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

@@ -11,14 +11,15 @@ from pathlib import Path
 from typing import Callable, Optional
 
 # Try to import readline with fallback for Windows
+HAS_READLINE = False
+readline = None  # type: ignore[assignment]
 try:
-    import readline
+    import readline  # type: ignore[no-redef]
 
     HAS_READLINE = True
 except ImportError:
     # Windows doesn't have readline by default
-    HAS_READLINE = False
+    pass
-    readline = None
 
 
 class InteractiveSession:

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

@@ -7,7 +7,7 @@ operators for different data types including numbers, strings, booleans, and lis
 """
 
 import logging
-from typing import Any, Union
+from typing import Any, Optional, Union
 
 logger = logging.getLogger(__name__)
 
@@ -47,7 +47,7 @@ class MetadataFilterEngine:
         }
 
     def apply_filters(
-        self, search_results: list[dict[str, Any]], metadata_filters: MetadataFilters
+        self, search_results: list[dict[str, Any]], metadata_filters: Optional[MetadataFilters]
     ) -> list[dict[str, Any]]:
         """
         Apply metadata filters to a list of search results.

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

@@ -56,7 +56,9 @@ class BaseSearcher(LeannBackendSearcherInterface, ABC):
         with open(meta_path, encoding="utf-8") as f:
             return json.load(f)
 
-    def _ensure_server_running(self, passages_source_file: str, port: int, **kwargs) -> int:
+    def _ensure_server_running(
+        self, passages_source_file: str, port: Optional[int], **kwargs
+    ) -> int:
         """
         Ensures the embedding server is running if recompute is needed.
         This is a helper for subclasses.
@@ -81,7 +83,7 @@ class BaseSearcher(LeannBackendSearcherInterface, ABC):
         }
 
         server_started, actual_port = self.embedding_server_manager.start_server(
-            port=port,
+            port=port if port is not None else 5557,
             model_name=self.embedding_model,
             embedding_mode=self.embedding_mode,
             passages_file=passages_source_file,
@@ -98,7 +100,7 @@ class BaseSearcher(LeannBackendSearcherInterface, ABC):
         self,
         query: str,
         use_server_if_available: bool = True,
-        zmq_port: int = 5557,
+        zmq_port: Optional[int] = None,
         query_template: Optional[str] = None,
     ) -> np.ndarray:
         """

packages/leann/pyproject.toml

@@ -7,7 +7,7 @@ name = "leann"
 version = "0.3.5"
 description = "LEANN - The smallest vector index in the world. RAG Everything with LEANN!"
 readme = "README.md"
-requires-python = ">=3.9"
+requires-python = ">=3.10"
 license = { text = "MIT" }
 authors = [
     { name = "LEANN Team" }
@@ -18,10 +18,10 @@ classifiers = [
     "Intended Audience :: Developers",
     "License :: OSI Approved :: MIT License",
     "Programming Language :: Python :: 3",
-    "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
 ]
 
 # Default installation: core + hnsw + diskann

pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "setuptools.build_meta"
 [project]
 name = "leann-workspace"
 version = "0.1.0"
-requires-python = ">=3.9"
+requires-python = ">=3.10"
 
 dependencies = [
     "leann-core",
@@ -157,6 +157,19 @@ exclude = ["localhost", "127.0.0.1", "example.com"]
 exclude_path = [".git/", ".venv/", "__pycache__/", "third_party/"]
 scheme = ["https", "http"]
 
+[tool.ty]
+# Type checking with ty (Astral's fast Python type checker)
+# ty is 10-100x faster than mypy. See: https://docs.astral.sh/ty/
+
+[tool.ty.environment]
+python-version = "3.11"
+extra-paths = ["apps", "packages/leann-core/src"]
+
+[tool.ty.rules]
+# Disable some noisy rules that have many false positives
+possibly-missing-attribute = "ignore"
+unresolved-import = "ignore"  # Many optional dependencies
+
 [tool.pytest.ini_options]
 testpaths = ["tests"]
 python_files = ["test_*.py"]

tests/test_basic.py

@@ -91,7 +91,7 @@ def test_large_index():
         builder.build_index(index_path)
 
         searcher = LeannSearcher(index_path)
-        results = searcher.search(["word10 word20"], top_k=10)
+        results = searcher.search("word10 word20", top_k=10)
-        assert len(results[0]) == 10
+        assert len(results) == 10
         # Cleanup
         searcher.cleanup()

tests/test_cli_prompt_template.py

@@ -123,7 +123,7 @@ class TestPromptTemplateStoredInEmbeddingOptions:
         cli = LeannCLI()
 
         # Mock load_documents to return a document so builder is created
-        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])
+        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])  # type: ignore[assignment]
 
         parser = cli.create_parser()
 
@@ -175,7 +175,7 @@ class TestPromptTemplateStoredInEmbeddingOptions:
         cli = LeannCLI()
 
         # Mock load_documents to return a document so builder is created
-        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])
+        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])  # type: ignore[assignment]
 
         parser = cli.create_parser()
 
@@ -230,7 +230,7 @@ class TestPromptTemplateStoredInEmbeddingOptions:
         cli = LeannCLI()
 
         # Mock load_documents to return a document so builder is created
-        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])
+        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])  # type: ignore[assignment]
 
         parser = cli.create_parser()
 
@@ -307,7 +307,7 @@ class TestPromptTemplateStoredInEmbeddingOptions:
         cli = LeannCLI()
 
         # Mock load_documents to return a document so builder is created
-        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])
+        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])  # type: ignore[assignment]
 
         parser = cli.create_parser()
 
@@ -376,7 +376,7 @@ class TestPromptTemplateStoredInEmbeddingOptions:
         cli = LeannCLI()
 
         # Mock load_documents to return a document so builder is created
-        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])
+        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])  # type: ignore[assignment]
 
         parser = cli.create_parser()
 
@@ -432,7 +432,7 @@ class TestPromptTemplateFlowsToComputeEmbeddings:
         cli = LeannCLI()
 
         # Mock load_documents to return a simple document
-        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])
+        cli.load_documents = Mock(return_value=[{"text": "test content", "metadata": {}}])  # type: ignore[assignment]
 
         parser = cli.create_parser()
 

tests/test_prompt_template_e2e.py

@@ -67,7 +67,7 @@ def check_lmstudio_available() -> bool:
         return False
 
 
-def get_lmstudio_first_model() -> str:
+def get_lmstudio_first_model() -> str | None:
     """Get the first available model from LM Studio."""
     try:
         response = requests.get("http://localhost:1234/v1/models", timeout=5.0)
@@ -91,6 +91,7 @@ class TestPromptTemplateOpenAI:
         model_name = get_lmstudio_first_model()
         if not model_name:
             pytest.skip("No models loaded in LM Studio")
+        assert model_name is not None  # Type narrowing for type checker
 
         texts = ["artificial intelligence", "machine learning"]
         prompt_template = "search_query: "
@@ -120,6 +121,7 @@ class TestPromptTemplateOpenAI:
         model_name = get_lmstudio_first_model()
         if not model_name:
             pytest.skip("No models loaded in LM Studio")
+        assert model_name is not None  # Type narrowing for type checker
 
         text = "machine learning"
         base_url = "http://localhost:1234/v1"
@@ -271,6 +273,7 @@ class TestLMStudioSDK:
         model_name = get_lmstudio_first_model()
         if not model_name:
             pytest.skip("No models loaded in LM Studio")
+        assert model_name is not None  # Type narrowing for type checker
 
         try:
             from leann.embedding_compute import _query_lmstudio_context_limit

tests/test_prompt_template_persistence.py

@@ -581,7 +581,18 @@ class TestQueryTemplateApplicationInComputeEmbedding:
 
         # Create a concrete implementation for testing
         class TestSearcher(BaseSearcher):
-            def search(self, query_vectors, top_k, complexity, beam_width=1, **kwargs):
+            def search(
+                self,
+                query,
+                top_k,
+                complexity=64,
+                beam_width=1,
+                prune_ratio=0.0,
+                recompute_embeddings=False,
+                pruning_strategy="global",
+                zmq_port=None,
+                **kwargs,
+            ):
                 return {"labels": [], "distances": []}
 
         searcher = object.__new__(TestSearcher)
@@ -625,7 +636,18 @@ class TestQueryTemplateApplicationInComputeEmbedding:
 
         # Create a concrete implementation for testing
         class TestSearcher(BaseSearcher):
-            def search(self, query_vectors, top_k, complexity, beam_width=1, **kwargs):
+            def search(
+                self,
+                query,
+                top_k,
+                complexity=64,
+                beam_width=1,
+                prune_ratio=0.0,
+                recompute_embeddings=False,
+                pruning_strategy="global",
+                zmq_port=None,
+                **kwargs,
+            ):
                 return {"labels": [], "distances": []}
 
         searcher = object.__new__(TestSearcher)
@@ -671,7 +693,18 @@ class TestQueryTemplateApplicationInComputeEmbedding:
         from leann.searcher_base import BaseSearcher
 
         class TestSearcher(BaseSearcher):
-            def search(self, query_vectors, top_k, complexity, beam_width=1, **kwargs):
+            def search(
+                self,
+                query,
+                top_k,
+                complexity=64,
+                beam_width=1,
+                prune_ratio=0.0,
+                recompute_embeddings=False,
+                pruning_strategy="global",
+                zmq_port=None,
+                **kwargs,
+            ):
                 return {"labels": [], "distances": []}
 
         searcher = object.__new__(TestSearcher)
@@ -710,7 +743,18 @@ class TestQueryTemplateApplicationInComputeEmbedding:
         from leann.searcher_base import BaseSearcher
 
         class TestSearcher(BaseSearcher):
-            def search(self, query_vectors, top_k, complexity, beam_width=1, **kwargs):
+            def search(
+                self,
+                query,
+                top_k,
+                complexity=64,
+                beam_width=1,
+                prune_ratio=0.0,
+                recompute_embeddings=False,
+                pruning_strategy="global",
+                zmq_port=None,
+                **kwargs,
+            ):
                 return {"labels": [], "distances": []}
 
         searcher = object.__new__(TestSearcher)
@@ -774,7 +818,18 @@ class TestQueryTemplateApplicationInComputeEmbedding:
         from leann.searcher_base import BaseSearcher
 
         class TestSearcher(BaseSearcher):
-            def search(self, query_vectors, top_k, complexity, beam_width=1, **kwargs):
+            def search(
+                self,
+                query,
+                top_k,
+                complexity=64,
+                beam_width=1,
+                prune_ratio=0.0,
+                recompute_embeddings=False,
+                pruning_strategy="global",
+                zmq_port=None,
+                **kwargs,
+            ):
                 return {"labels": [], "distances": []}
 
         searcher = object.__new__(TestSearcher)

tests/test_readme_examples.py

@@ -97,17 +97,17 @@ def test_backend_options():
 
     with tempfile.TemporaryDirectory() as temp_dir:
         # Use smaller model in CI to avoid memory issues
-        if os.environ.get("CI") == "true":
+        is_ci = os.environ.get("CI") == "true"
-            model_args = {
+        embedding_model = (
-                "embedding_model": "sentence-transformers/all-MiniLM-L6-v2",
+            "sentence-transformers/all-MiniLM-L6-v2" if is_ci else "facebook/contriever"
-                "dimensions": 384,
+        )
-            }
+        dimensions = 384 if is_ci else None
-        else:
-            model_args = {}
 
         # Test HNSW backend (as shown in README)
         hnsw_path = str(Path(temp_dir) / "test_hnsw.leann")
-        builder_hnsw = LeannBuilder(backend_name="hnsw", **model_args)
+        builder_hnsw = LeannBuilder(
+            backend_name="hnsw", embedding_model=embedding_model, dimensions=dimensions
+        )
         builder_hnsw.add_text("Test document for HNSW backend")
         builder_hnsw.build_index(hnsw_path)
         assert Path(hnsw_path).parent.exists()
@@ -115,7 +115,9 @@ def test_backend_options():
 
         # Test DiskANN backend (mentioned as available option)
         diskann_path = str(Path(temp_dir) / "test_diskann.leann")
-        builder_diskann = LeannBuilder(backend_name="diskann", **model_args)
+        builder_diskann = LeannBuilder(
+            backend_name="diskann", embedding_model=embedding_model, dimensions=dimensions
+        )
         builder_diskann.add_text("Test document for DiskANN backend")
         builder_diskann.build_index(diskann_path)
         assert Path(diskann_path).parent.exists()