diff --git a/benchmarks/financebench/evaluate_financebench.py b/benchmarks/financebench/evaluate_financebench.py
index 4d3370c..ac05cce 100755
--- a/benchmarks/financebench/evaluate_financebench.py
+++ b/benchmarks/financebench/evaluate_financebench.py
@@ -1,19 +1,108 @@
-#!/usr/bin/env python3
 """
-FinanceBench Evaluation Script
-Uses intelligent evaluation similar to VectifyAI/Mafin approach
+FinanceBench Evaluation Script - Modular Recall-based Evaluation
 """
 
 import argparse
 import json
 import os
-import re
+import pickle
 import time
 from typing import Optional
 
+import numpy as np
 import openai
+
+# Import LEANN modules - this will bring in the modified faiss
 from leann import LeannChat, LeannSearcher
 
+# Import LEANN's modified faiss directly
+from leann_backend_hnsw import faiss
+
+
+class RecallEvaluator:
+    """Stage 2: Evaluate Recall@3 (searcher vs baseline)"""
+
+    def __init__(self, index_path: str, baseline_dir: str):
+        self.index_path = index_path
+        self.baseline_dir = baseline_dir
+        self.searcher = LeannSearcher(index_path)
+
+        # Load FAISS flat baseline
+        baseline_index_path = os.path.join(baseline_dir, "faiss_flat.index")
+        metadata_path = os.path.join(baseline_dir, "metadata.pkl")
+
+        self.faiss_index = faiss.read_index(baseline_index_path)
+        with open(metadata_path, "rb") as f:
+            self.passage_ids = pickle.load(f)
+        print(f"📚 Loaded FAISS flat baseline with {self.faiss_index.ntotal} vectors")
+
+    def evaluate_recall_at_3(
+        self, queries: list[str], complexity: int = 64, recompute_embeddings: bool = True
+    ) -> float:
+        """Evaluate recall@3 for given queries at specified complexity"""
+        recompute_str = "with recompute" if recompute_embeddings else "no recompute"
+        print(f"🔍 Evaluating recall@3 with complexity={complexity} ({recompute_str})...")
+
+        total_recall = 0.0
+        num_queries = len(queries)
+
+        for i, query in enumerate(queries):
+            # Get ground truth: search with FAISS flat
+            from leann.api import compute_embeddings
+
+            query_embedding = compute_embeddings(
+                [query],
+                self.searcher.embedding_model,
+                mode=self.searcher.embedding_mode,
+                use_server=False,
+            ).astype(np.float32)
+
+            # Search FAISS flat for ground truth using LEANN's modified faiss API
+            n = query_embedding.shape[0]  # Number of queries
+            k = 3  # Number of nearest neighbors
+            distances = np.zeros((n, k), dtype=np.float32)
+            labels = np.zeros((n, k), dtype=np.int64)
+
+            self.faiss_index.search(
+                n,
+                faiss.swig_ptr(query_embedding),
+                k,
+                faiss.swig_ptr(distances),
+                faiss.swig_ptr(labels),
+            )
+
+            # Extract the results
+            baseline_ids = {self.passage_ids[idx] for idx in labels[0]}
+
+            # Search with LEANN at specified complexity
+            test_results = self.searcher.search(
+                query,
+                top_k=3,
+                complexity=complexity,
+                recompute_embeddings=recompute_embeddings,
+            )
+            test_ids = {result.id for result in test_results}
+
+            # Calculate recall@3 = |intersection| / |ground_truth|
+            intersection = test_ids.intersection(baseline_ids)
+            recall = len(intersection) / 3.0  # Ground truth size is 3
+            total_recall += recall
+
+            if i < 3:  # Show first few examples
+                print(f"  Query {i + 1}: '{query[:50]}...' -> Recall@3: {recall:.3f}")
+                print(f"    FAISS ground truth: {list(baseline_ids)}")
+                print(f"    LEANN results (C={complexity}, {recompute_str}): {list(test_ids)}")
+                print(f"    Intersection: {list(intersection)}")
+
+        avg_recall = total_recall / num_queries
+        print(f"📊 Average Recall@3: {avg_recall:.3f} ({avg_recall * 100:.1f}%)")
+        return avg_recall
+
+    def cleanup(self):
+        """Cleanup resources"""
+        if hasattr(self, "searcher"):
+            self.searcher.cleanup()
+
 
 class FinanceBenchEvaluator:
     def __init__(self, index_path: str, openai_api_key: Optional[str] = None):
@@ -34,310 +123,472 @@ class FinanceBenchEvaluator:
         print(f"📊 Loaded {len(data)} FinanceBench examples")
         return data
 
-    def evaluate_retrieval_intelligent(self, data: list[dict], top_k: int = 10) -> dict:
-        """
-        Intelligent retrieval evaluation
-        Uses semantic similarity instead of strict word overlap
-        """
-        print(f"🔍 Evaluating retrieval performance (top_k={top_k})...")
+    def analyze_index_sizes(self) -> dict:
+        """Analyze index sizes with and without embeddings"""
+        from pathlib import Path
 
-        metrics = {
-            "total_questions": 0,
-            "questions_with_relevant_retrieved": 0,
-            "exact_matches": 0,
-            "semantic_matches": 0,
-            "number_matches": 0,
-            "search_times": [],
-            "detailed_results": [],
+        print("📏 Analyzing index sizes...")
+
+        # Get all index-related files
+        index_path = Path(self.index_path)
+        index_dir = index_path.parent
+        index_name = index_path.stem  # Remove .leann extension
+
+        sizes = {}
+        total_with_embeddings = 0
+        total_without_embeddings = 0
+
+        # Core index files
+        index_file = index_dir / f"{index_name}.index"
+        meta_file = index_dir / f"{index_path.name}.meta.json"  # Keep .leann for meta file
+        passages_file = index_dir / f"{index_path.name}.passages.jsonl"  # Keep .leann for passages
+        passages_idx_file = index_dir / f"{index_path.name}.passages.idx"  # Keep .leann for idx
+
+        for file_path, name in [
+            (index_file, "index"),
+            (meta_file, "metadata"),
+            (passages_file, "passages_text"),
+            (passages_idx_file, "passages_index"),
+        ]:
+            if file_path.exists():
+                size_mb = file_path.stat().st_size / (1024 * 1024)
+                sizes[name] = size_mb
+                total_with_embeddings += size_mb
+
+                # For pruned index calculation, exclude the main index file (contains embeddings)
+                if name != "index":
+                    total_without_embeddings += size_mb
+            else:
+                sizes[name] = 0
+
+        # Estimate pruned index size (approximate)
+        # When embeddings are removed, the main index file becomes much smaller
+        # Rough estimate: graph structure is ~10-20% of full index size
+        estimated_pruned_index_size = sizes["index"] * 0.15  # Conservative estimate
+        total_without_embeddings += estimated_pruned_index_size
+
+        sizes["total_with_embeddings"] = total_with_embeddings
+        sizes["total_without_embeddings"] = total_without_embeddings
+        sizes["estimated_pruned_index"] = estimated_pruned_index_size
+        sizes["compression_ratio"] = (
+            total_without_embeddings / total_with_embeddings if total_with_embeddings > 0 else 0
+        )
+
+        print(f"  📁 Index with embeddings: {total_with_embeddings:.1f} MB")
+        print(f"  📁 Estimated pruned index: {total_without_embeddings:.1f} MB")
+        print(f"  🗜️  Compression ratio: {sizes['compression_ratio']:.2f}x")
+
+        return sizes
+
+    def create_compact_index_for_comparison(self, compact_index_path: str) -> dict:
+        """Create a compact index for comparison purposes"""
+        print("🏗️ Building compact index from existing passages...")
+
+        # Load existing passages from current index
+        from pathlib import Path
+
+        from leann import LeannBuilder
+
+        current_index_path = Path(self.index_path)
+        current_index_dir = current_index_path.parent
+        current_index_name = current_index_path.name
+
+        # Read metadata to get passage source
+        meta_path = current_index_dir / f"{current_index_name}.meta.json"
+        with open(meta_path) as f:
+            import json
+
+            meta = json.load(f)
+
+        passage_source = meta["passage_sources"][0]
+        passage_file = passage_source["path"]
+
+        # Convert relative path to absolute
+        if not Path(passage_file).is_absolute():
+            passage_file = current_index_dir / Path(passage_file).name
+
+        print(f"📄 Loading passages from {passage_file}...")
+
+        # Build compact index with same passages
+        builder = LeannBuilder(
+            backend_name="hnsw",
+            embedding_model=meta["embedding_model"],
+            embedding_mode=meta.get("embedding_mode", "sentence-transformers"),
+            is_recompute=True,  # Enable recompute (no stored embeddings)
+            is_compact=True,  # Enable compact storage
+            **meta.get("backend_kwargs", {}),
+        )
+
+        # Load all passages
+        with open(passage_file, encoding="utf-8") as f:
+            for line in f:
+                if line.strip():
+                    data = json.loads(line)
+                    builder.add_text(data["text"], metadata=data.get("metadata", {}))
+
+        print(f"🔨 Building compact index at {compact_index_path}...")
+        builder.build_index(compact_index_path)
+
+        # Analyze the compact index size
+        temp_evaluator = FinanceBenchEvaluator(compact_index_path)
+        compact_sizes = temp_evaluator.analyze_index_sizes()
+        compact_sizes["index_type"] = "compact"
+
+        return compact_sizes
+
+    def create_non_compact_index_for_comparison(self, non_compact_index_path: str) -> dict:
+        """Create a non-compact index for comparison purposes"""
+        print("🏗️ Building non-compact index from existing passages...")
+
+        # Load existing passages from current index
+        from pathlib import Path
+
+        from leann import LeannBuilder
+
+        current_index_path = Path(self.index_path)
+        current_index_dir = current_index_path.parent
+        current_index_name = current_index_path.name
+
+        # Read metadata to get passage source
+        meta_path = current_index_dir / f"{current_index_name}.meta.json"
+        with open(meta_path) as f:
+            import json
+
+            meta = json.load(f)
+
+        passage_source = meta["passage_sources"][0]
+        passage_file = passage_source["path"]
+
+        # Convert relative path to absolute
+        if not Path(passage_file).is_absolute():
+            passage_file = current_index_dir / Path(passage_file).name
+
+        print(f"📄 Loading passages from {passage_file}...")
+
+        # Build non-compact index with same passages
+        builder = LeannBuilder(
+            backend_name="hnsw",
+            embedding_model=meta["embedding_model"],
+            embedding_mode=meta.get("embedding_mode", "sentence-transformers"),
+            is_recompute=False,  # Disable recompute (store embeddings)
+            is_compact=False,  # Disable compact storage
+            **{
+                k: v
+                for k, v in meta.get("backend_kwargs", {}).items()
+                if k not in ["is_recompute", "is_compact"]
+            },
+        )
+
+        # Load all passages
+        with open(passage_file, encoding="utf-8") as f:
+            for line in f:
+                if line.strip():
+                    data = json.loads(line)
+                    builder.add_text(data["text"], metadata=data.get("metadata", {}))
+
+        print(f"🔨 Building non-compact index at {non_compact_index_path}...")
+        builder.build_index(non_compact_index_path)
+
+        # Analyze the non-compact index size
+        temp_evaluator = FinanceBenchEvaluator(non_compact_index_path)
+        non_compact_sizes = temp_evaluator.analyze_index_sizes()
+        non_compact_sizes["index_type"] = "non_compact"
+
+        return non_compact_sizes
+
+    def compare_index_performance(
+        self, non_compact_path: str, compact_path: str, test_data: list, complexity: int
+    ) -> dict:
+        """Compare performance between non-compact and compact indexes"""
+        print("⚡ Comparing search performance between indexes...")
+
+        import time
+
+        from leann import LeannSearcher
+
+        # Test queries
+        test_queries = [item["question"] for item in test_data[:5]]
+
+        results = {
+            "non_compact": {"search_times": []},
+            "compact": {"search_times": []},
+            "avg_search_times": {},
+            "speed_ratio": 0.0,
         }
 
-        for item in data:
-            question = item["question"]
-            evidence_texts = [ev["evidence_text"] for ev in item.get("evidence", [])]
-            expected_answer = item["answer"]
+        # Test non-compact index (no recompute)
+        print("  🔍 Testing non-compact index (no recompute)...")
+        non_compact_searcher = LeannSearcher(non_compact_path)
 
-            if not evidence_texts:
-                continue
-
-            metrics["total_questions"] += 1
-
-            # Search for relevant documents
+        for query in test_queries:
             start_time = time.time()
-            results = self.searcher.search(question, top_k=top_k, complexity=64)
+            search_results = non_compact_searcher.search(
+                query, top_k=3, complexity=complexity, recompute_embeddings=False
+            )
             search_time = time.time() - start_time
-            metrics["search_times"].append(search_time)
+            results["non_compact"]["search_times"].append(search_time)
 
-            # Evaluate retrieved results
-            found_relevant = False
-            match_types = []
+        # Test compact index (with recompute)
+        print("  🔍 Testing compact index (with recompute)...")
+        compact_searcher = LeannSearcher(compact_path)
 
-            for evidence_text in evidence_texts:
-                for i, result in enumerate(results):
-                    retrieved_text = result.text
-
-                    # Method 1: Exact substring match
-                    if self._has_exact_overlap(evidence_text, retrieved_text):
-                        found_relevant = True
-                        match_types.append(f"exact_match_rank_{i + 1}")
-                        metrics["exact_matches"] += 1
-                        break
-
-                    # Method 2: Key numbers match
-                    elif self._has_number_match(evidence_text, retrieved_text, expected_answer):
-                        found_relevant = True
-                        match_types.append(f"number_match_rank_{i + 1}")
-                        metrics["number_matches"] += 1
-                        break
-
-                    # Method 3: Semantic similarity (word overlap with lower threshold)
-                    elif self._has_semantic_similarity(
-                        evidence_text, retrieved_text, threshold=0.2
-                    ):
-                        found_relevant = True
-                        match_types.append(f"semantic_match_rank_{i + 1}")
-                        metrics["semantic_matches"] += 1
-                        break
-
-            if found_relevant:
-                metrics["questions_with_relevant_retrieved"] += 1
-
-            # Store detailed result
-            metrics["detailed_results"].append(
-                {
-                    "question": question,
-                    "expected_answer": expected_answer,
-                    "found_relevant": found_relevant,
-                    "match_types": match_types,
-                    "search_time": search_time,
-                    "top_results": [
-                        {"text": r.text[:200] + "...", "score": r.score, "metadata": r.metadata}
-                        for r in results[:3]
-                    ],
-                }
+        for query in test_queries:
+            start_time = time.time()
+            search_results = compact_searcher.search(
+                query, top_k=3, complexity=complexity, recompute_embeddings=True
             )
+            search_time = time.time() - start_time
+            results["compact"]["search_times"].append(search_time)
 
-        # Calculate metrics
-        if metrics["total_questions"] > 0:
-            metrics["question_coverage"] = (
-                metrics["questions_with_relevant_retrieved"] / metrics["total_questions"]
+        # Calculate averages
+        results["avg_search_times"]["non_compact"] = sum(
+            results["non_compact"]["search_times"]
+        ) / len(results["non_compact"]["search_times"])
+        results["avg_search_times"]["compact"] = sum(results["compact"]["search_times"]) / len(
+            results["compact"]["search_times"]
+        )
+
+        # Performance ratio
+        if results["avg_search_times"]["compact"] > 0:
+            results["speed_ratio"] = (
+                results["avg_search_times"]["non_compact"] / results["avg_search_times"]["compact"]
             )
-            metrics["avg_search_time"] = sum(metrics["search_times"]) / len(metrics["search_times"])
+        else:
+            results["speed_ratio"] = float("inf")
 
-            # Match type breakdown
-            metrics["exact_match_rate"] = metrics["exact_matches"] / metrics["total_questions"]
-            metrics["number_match_rate"] = metrics["number_matches"] / metrics["total_questions"]
-            metrics["semantic_match_rate"] = (
-                metrics["semantic_matches"] / metrics["total_questions"]
-            )
+        print(
+            f"    Non-compact (no recompute): {results['avg_search_times']['non_compact']:.3f}s avg"
+        )
+        print(f"    Compact (with recompute): {results['avg_search_times']['compact']:.3f}s avg")
+        print(f"    Speed ratio: {results['speed_ratio']:.2f}x")
 
-        return metrics
+        # Cleanup
+        non_compact_searcher.cleanup()
+        compact_searcher.cleanup()
 
-    def evaluate_qa_intelligent(self, data: list[dict], max_samples: Optional[int] = None) -> dict:
-        """
-        Intelligent QA evaluation using LLM-based answer comparison
-        Similar to VectifyAI/Mafin approach
-        """
+        return results
+
+    def evaluate_timing_breakdown(
+        self, data: list[dict], max_samples: Optional[int] = None
+    ) -> dict:
+        """Evaluate timing breakdown and accuracy by hacking LeannChat.ask() for separated timing"""
         if not self.chat or not self.openai_client:
-            print("⚠️  Skipping QA evaluation (no OpenAI API key provided)")
-            return {"accuracy": 0.0, "total_questions": 0}
+            print("⚠️  Skipping timing evaluation (no OpenAI API key provided)")
+            return {
+                "total_questions": 0,
+                "avg_search_time": 0.0,
+                "avg_generation_time": 0.0,
+                "avg_total_time": 0.0,
+                "accuracy": 0.0,
+            }
 
-        print("🤖 Evaluating QA performance...")
+        print("🔍🤖 Evaluating timing breakdown and accuracy (search + generation)...")
 
         if max_samples:
             data = data[:max_samples]
-            print(f"📝 Using first {max_samples} samples for QA evaluation")
+            print(f"📝 Using first {max_samples} samples for timing evaluation")
 
-        results = []
+        search_times = []
+        generation_times = []
+        total_times = []
         correct_answers = 0
 
         for i, item in enumerate(data):
             question = item["question"]
-            expected_answer = item["answer"]
-
-            print(f"Question {i + 1}/{len(data)}: {question[:80]}...")
+            ground_truth = item["answer"]
 
             try:
-                # Get answer from LEANN
-                start_time = time.time()
+                # Hack: Monkey-patch the ask method to capture internal timing
+                original_ask = self.chat.ask
+                captured_search_time = None
+                captured_generation_time = None
+
+                def patched_ask(*args, **kwargs):
+                    nonlocal captured_search_time, captured_generation_time
+
+                    # Time the search part
+                    search_start = time.time()
+                    results = self.chat.searcher.search(args[0], top_k=3, complexity=64)
+                    captured_search_time = time.time() - search_start
+
+                    # Time the generation part
+                    context = "\n\n".join([r.text for r in results])
+                    prompt = (
+                        "Here is some retrieved context that might help answer your question:\n\n"
+                        f"{context}\n\n"
+                        f"Question: {args[0]}\n\n"
+                        "Please provide the best answer you can based on this context and your knowledge."
+                    )
+
+                    generation_start = time.time()
+                    answer = self.chat.llm.ask(prompt)
+                    captured_generation_time = time.time() - generation_start
+
+                    return answer
+
+                # Apply the patch
+                self.chat.ask = patched_ask
+
+                # Time the total QA
+                total_start = time.time()
                 generated_answer = self.chat.ask(question)
-                qa_time = time.time() - start_time
+                total_time = time.time() - total_start
 
-                # Intelligent evaluation using LLM
-                is_correct = self._evaluate_answer_with_llm(
-                    question, expected_answer, generated_answer
-                )
+                # Restore original method
+                self.chat.ask = original_ask
 
+                # Store the timings
+                search_times.append(captured_search_time)
+                generation_times.append(captured_generation_time)
+                total_times.append(total_time)
+
+                # Check accuracy using LLM as judge
+                is_correct = self._check_answer_accuracy(generated_answer, ground_truth, question)
                 if is_correct:
                     correct_answers += 1
 
-                results.append(
-                    {
-                        "question": question,
-                        "expected_answer": expected_answer,
-                        "generated_answer": generated_answer,
-                        "is_correct": is_correct,
-                        "qa_time": qa_time,
-                    }
+                status = "✅" if is_correct else "❌"
+                print(
+                    f"Question {i + 1}/{len(data)}: {status} Search={captured_search_time:.3f}s, Gen={captured_generation_time:.3f}s, Total={total_time:.3f}s"
                 )
-
-                print(f"  ✅ {'Correct' if is_correct else '❌ Incorrect'}")
+                print(f"  GT: {ground_truth}")
+                print(f"  Gen: {generated_answer[:100]}...")
 
             except Exception as e:
                 print(f"  ❌ Error: {e}")
-                results.append(
-                    {
-                        "question": question,
-                        "expected_answer": expected_answer,
-                        "generated_answer": f"ERROR: {e}",
-                        "is_correct": False,
-                        "qa_time": 0.0,
-                    }
-                )
+                search_times.append(0.0)
+                generation_times.append(0.0)
+                total_times.append(0.0)
+
+        accuracy = correct_answers / len(data) if data else 0.0
 
         metrics = {
             "total_questions": len(data),
+            "avg_search_time": sum(search_times) / len(search_times) if search_times else 0.0,
+            "avg_generation_time": sum(generation_times) / len(generation_times)
+            if generation_times
+            else 0.0,
+            "avg_total_time": sum(total_times) / len(total_times) if total_times else 0.0,
+            "accuracy": accuracy,
             "correct_answers": correct_answers,
-            "accuracy": correct_answers / len(data) if data else 0.0,
-            "avg_qa_time": sum(r["qa_time"] for r in results) / len(results) if results else 0.0,
-            "detailed_results": results,
+            "search_times": search_times,
+            "generation_times": generation_times,
+            "total_times": total_times,
         }
 
         return metrics
 
-    def _has_exact_overlap(self, evidence_text: str, retrieved_text: str) -> bool:
-        """Check for exact substring overlap"""
-        # Check if evidence is contained in retrieved text or vice versa
-        return (
-            evidence_text.lower() in retrieved_text.lower()
-            or retrieved_text.lower() in evidence_text.lower()
-        )
-
-    def _has_number_match(
-        self, evidence_text: str, retrieved_text: str, expected_answer: str
+    def _check_answer_accuracy(
+        self, generated_answer: str, ground_truth: str, question: str
     ) -> bool:
-        """Check if key numbers from evidence/answer appear in retrieved text"""
-        # Extract numbers from evidence and expected answer
-        evidence_numbers = set(re.findall(r"\$?[\d,]+\.?\d*", evidence_text))
-        answer_numbers = set(re.findall(r"\$?[\d,]+\.?\d*", expected_answer))
-        retrieved_numbers = set(re.findall(r"\$?[\d,]+\.?\d*", retrieved_text))
-
-        # Check if any key numbers match
-        key_numbers = evidence_numbers.union(answer_numbers)
-        return bool(key_numbers.intersection(retrieved_numbers))
-
-    def _has_semantic_similarity(
-        self, evidence_text: str, retrieved_text: str, threshold: float = 0.2
-    ) -> bool:
-        """Check semantic similarity using word overlap"""
-        words1 = set(evidence_text.lower().split())
-        words2 = set(retrieved_text.lower().split())
-
-        if len(words1) == 0:
-            return False
-
-        overlap = len(words1.intersection(words2))
-        similarity = overlap / len(words1)
-
-        return similarity >= threshold
-
-    def _evaluate_answer_with_llm(
-        self, question: str, expected_answer: str, generated_answer: str
-    ) -> bool:
-        """
-        Use LLM to evaluate answer equivalence
-        Based on VectifyAI/Mafin approach
-        """
-        prompt = f"""You are an expert evaluator for AI-generated responses to financial questions. Your task is to determine whether the AI-generated answer correctly answers the question based on the golden answer provided by a human expert.
-
-Evaluation Criteria:
-- Numerical Accuracy: Rounding differences should be ignored if they don't meaningfully change the conclusion. Numbers like 1.2 and 1.23 are considered similar.
-- Fractions, percentages, and numerics could be considered similar. For example: "11 of 14" ≈ "79%" ≈ "0.79".
-- If the golden answer or any of its equivalence can be inferred from the AI answer, then the AI answer is correct.
-- The AI answer is correct if it conveys the same meaning, conclusion, or rationale as the golden answer.
-- If the AI answer is a superset of the golden answer, it is also considered correct.
-- Subjective judgments are correct as long as they are reasonable and justifiable.
+        """Check if generated answer matches ground truth using LLM as judge"""
+        judge_prompt = f"""You are an expert judge evaluating financial question answering. 
 
 Question: {question}
-AI-Generated Answer: {generated_answer}
-Golden Answer: {expected_answer}
 
-Your output should be ONLY a boolean value: `True` or `False`, nothing else."""
+Ground Truth Answer: {ground_truth}
 
-        # retry in exponential backoff
-        for i in range(3):
-            try:
-                response = self.openai_client.chat.completions.create(
-                    model="gpt-4o",
-                    messages=[{"role": "user", "content": prompt}],
-                    temperature=0,
-                    max_tokens=10,
-                )
-                result = response.choices[0].message.content.strip().lower()
-                return "true" in result
-            except Exception as e:
-                print(f"LLM evaluation error: {e}")
-                time.sleep(2**i)
+Generated Answer: {generated_answer}
 
-        return False
+Task: Determine if the generated answer is factually correct compared to the ground truth. Focus on:
+1. Numerical accuracy (exact values, units, currency)
+2. Key financial concepts and terminology
+3. Overall factual correctness
 
-    def run_evaluation(
-        self,
-        dataset_path: str = "data/financebench_merged.jsonl",
-        top_k: int = 10,
-        qa_samples: Optional[int] = None,
-    ) -> dict:
-        """Run complete FinanceBench evaluation"""
-        print("🏦 FinanceBench Evaluation with LEANN")
-        print("=" * 50)
-        print(f"📁 Index: {self.index_path}")
-        print(f"🔍 Top-k: {top_k}")
-        if qa_samples:
-            print(f"🤖 QA samples: {qa_samples}")
-        print()
+For financial data, small formatting differences are OK (e.g., "$1,577" vs "1577 million" vs "$1.577 billion"), but the core numerical value must match.
 
-        # Load dataset
-        data = self.load_dataset(dataset_path)
+Respond with exactly one word: "CORRECT" if the generated answer is factually accurate, or "INCORRECT" if it's wrong or significantly different."""
 
-        # Run retrieval evaluation
-        retrieval_metrics = self.evaluate_retrieval_intelligent(data, top_k=top_k)
+        try:
+            judge_response = self.openai_client.chat.completions.create(
+                model="gpt-4o-mini",
+                messages=[{"role": "user", "content": judge_prompt}],
+                max_tokens=10,
+                temperature=0,
+            )
+            judgment = judge_response.choices[0].message.content.strip().upper()
+            return judgment == "CORRECT"
+        except Exception as e:
+            print(f"  ⚠️  Judge error: {e}, falling back to string matching")
+            # Fallback to simple string matching
+            gen_clean = generated_answer.strip().lower().replace("$", "").replace(",", "")
+            gt_clean = ground_truth.strip().lower().replace("$", "").replace(",", "")
+            return gt_clean in gen_clean
 
-        # Run QA evaluation
-        qa_metrics = self.evaluate_qa_intelligent(data, max_samples=qa_samples)
-
-        # Print results
-        self._print_results(retrieval_metrics, qa_metrics)
-
-        return {
-            "retrieval": retrieval_metrics,
-            "qa": qa_metrics,
-        }
-
-    def _print_results(self, retrieval_metrics: dict, qa_metrics: dict):
+    def _print_results(self, timing_metrics: dict):
         """Print evaluation results"""
         print("\n🎯 EVALUATION RESULTS")
         print("=" * 50)
 
-        print("\n📊 Retrieval Metrics:")
-        print(f"  Question Coverage: {retrieval_metrics.get('question_coverage', 0):.1%}")
-        print(f"  Exact Match Rate: {retrieval_metrics.get('exact_match_rate', 0):.1%}")
-        print(f"  Number Match Rate: {retrieval_metrics.get('number_match_rate', 0):.1%}")
-        print(f"  Semantic Match Rate: {retrieval_metrics.get('semantic_match_rate', 0):.1%}")
-        print(f"  Avg Search Time: {retrieval_metrics.get('avg_search_time', 0):.3f}s")
+        # Index comparison analysis
+        if "current_index" in timing_metrics and "non_compact_index" in timing_metrics:
+            print("\n📏 Index Comparison Analysis:")
+            current = timing_metrics["current_index"]
+            non_compact = timing_metrics["non_compact_index"]
 
-        if qa_metrics.get("total_questions", 0) > 0:
-            print("\n🤖 QA Metrics:")
-            print(f"  Accuracy: {qa_metrics.get('accuracy', 0):.1%}")
-            print(f"  Questions Evaluated: {qa_metrics.get('total_questions', 0)}")
-            print(f"  Avg QA Time: {qa_metrics.get('avg_qa_time', 0):.3f}s")
+            print(f"  Compact index (current): {current.get('total_with_embeddings', 0):.1f} MB")
+            print(
+                f"  Non-compact index (with embeddings): {non_compact.get('total_with_embeddings', 0):.1f} MB"
+            )
+            print(
+                f"  Storage saving by compact: {timing_metrics.get('storage_saving_percent', 0):.1f}%"
+            )
 
-        # Show some example results
-        print("\n📝 Example Results:")
-        for i, result in enumerate(retrieval_metrics.get("detailed_results", [])[:3]):
-            print(f"\n  Example {i + 1}:")
-            print(f"    Q: {result['question'][:80]}...")
-            print(f"    Found relevant: {'✅' if result['found_relevant'] else '❌'}")
-            if result["match_types"]:
-                print(f"    Match types: {', '.join(result['match_types'])}")
+            print("  Component breakdown (non-compact):")
+            print(f"    - Main index: {non_compact.get('index', 0):.1f} MB")
+            print(f"    - Passages text: {non_compact.get('passages_text', 0):.1f} MB")
+            print(f"    - Passages index: {non_compact.get('passages_index', 0):.1f} MB")
+            print(f"    - Metadata: {non_compact.get('metadata', 0):.1f} MB")
+
+        # Performance comparison
+        if "performance_comparison" in timing_metrics:
+            perf = timing_metrics["performance_comparison"]
+            print("\n⚡ Performance Comparison:")
+            print(
+                f"  Non-compact (no recompute): {perf.get('avg_search_times', {}).get('non_compact', 0):.3f}s avg"
+            )
+            print(
+                f"  Compact (with recompute): {perf.get('avg_search_times', {}).get('compact', 0):.3f}s avg"
+            )
+            print(f"  Speed ratio: {perf.get('speed_ratio', 0):.2f}x")
+
+        # Legacy single index analysis (fallback)
+        if "total_with_embeddings" in timing_metrics and "current_index" not in timing_metrics:
+            print("\n📏 Index Size Analysis:")
+            print(
+                f"  Index with embeddings: {timing_metrics.get('total_with_embeddings', 0):.1f} MB"
+            )
+            print(
+                f"  Estimated pruned index: {timing_metrics.get('total_without_embeddings', 0):.1f} MB"
+            )
+            print(f"  Compression ratio: {timing_metrics.get('compression_ratio', 0):.2f}x")
+
+        print("\n📊 Accuracy:")
+        print(f"  Accuracy: {timing_metrics.get('accuracy', 0) * 100:.1f}%")
+        print(
+            f"  Correct Answers: {timing_metrics.get('correct_answers', 0)}/{timing_metrics.get('total_questions', 0)}"
+        )
+
+        print("\n📊 Timing Breakdown:")
+        print(f"  Total Questions: {timing_metrics.get('total_questions', 0)}")
+        print(f"  Avg Search Time: {timing_metrics.get('avg_search_time', 0):.3f}s")
+        print(f"  Avg Generation Time: {timing_metrics.get('avg_generation_time', 0):.3f}s")
+        print(f"  Avg Total Time: {timing_metrics.get('avg_total_time', 0):.3f}s")
+
+        if timing_metrics.get("avg_total_time", 0) > 0:
+            search_pct = (
+                timing_metrics.get("avg_search_time", 0)
+                / timing_metrics.get("avg_total_time", 1)
+                * 100
+            )
+            gen_pct = (
+                timing_metrics.get("avg_generation_time", 0)
+                / timing_metrics.get("avg_total_time", 1)
+                * 100
+            )
+            print("\n📈 Time Distribution:")
+            print(f"  Search: {search_pct:.1f}%")
+            print(f"  Generation: {gen_pct:.1f}%")
 
     def cleanup(self):
         """Cleanup resources"""
@@ -346,44 +597,299 @@ Your output should be ONLY a boolean value: `True` or `False`, nothing else."""
 
 
 def main():
-    parser = argparse.ArgumentParser(description="Evaluate FinanceBench with LEANN")
+    parser = argparse.ArgumentParser(description="Modular FinanceBench Evaluation")
     parser.add_argument("--index", required=True, help="Path to LEANN index")
     parser.add_argument("--dataset", default="data/financebench_merged.jsonl", help="Dataset path")
-    parser.add_argument("--top-k", type=int, default=10, help="Number of documents to retrieve")
-    parser.add_argument("--qa-samples", type=int, default=None, help="Limit QA evaluation samples")
-    parser.add_argument("--openai-api-key", help="OpenAI API key for QA evaluation")
+    parser.add_argument(
+        "--stage",
+        choices=["2", "3", "4", "all"],
+        default="all",
+        help="Which stage to run (2=recall, 3=complexity, 4=generation)",
+    )
+    parser.add_argument("--complexity", type=int, default=None, help="Complexity for search")
+    parser.add_argument("--baseline-dir", default="baseline", help="Baseline output directory")
+    parser.add_argument("--openai-api-key", help="OpenAI API key for generation evaluation")
     parser.add_argument("--output", help="Save results to JSON file")
 
     args = parser.parse_args()
 
-    # Get OpenAI API key
-    api_key = args.openai_api_key or os.getenv("OPENAI_API_KEY")
-    if not api_key and args.qa_samples != 0:
-        print("⚠️  No OpenAI API key provided. QA evaluation will be skipped.")
-        print("   Set OPENAI_API_KEY environment variable or use --openai-api-key")
-
     try:
-        # Run evaluation
-        evaluator = FinanceBenchEvaluator(args.index, api_key)
-        results = evaluator.run_evaluation(
-            dataset_path=args.dataset, top_k=args.top_k, qa_samples=args.qa_samples
-        )
+        # Check if baseline exists
+        baseline_index_path = os.path.join(args.baseline_dir, "faiss_flat.index")
+        if not os.path.exists(baseline_index_path):
+            print(f"❌ FAISS baseline not found at {baseline_index_path}")
+            print("💡 Please run setup_financebench.py first to build the baseline")
+            exit(1)
 
-        # Save results if requested
-        if args.output:
-            with open(args.output, "w") as f:
-                json.dump(results, f, indent=2, default=str)
-            print(f"\n💾 Results saved to {args.output}")
+        if args.stage == "2" or args.stage == "all":
+            # Stage 2: Recall@3 evaluation
+            print("🚀 Starting Stage 2: Recall@3 evaluation")
 
-        evaluator.cleanup()
+            evaluator = RecallEvaluator(args.index, args.baseline_dir)
 
-        print("\n✅ Evaluation completed!")
+            # Load FinanceBench queries for testing
+            print("📖 Loading FinanceBench dataset...")
+            queries = []
+            with open(args.dataset, encoding="utf-8") as f:
+                for line in f:
+                    if line.strip():
+                        data = json.loads(line)
+                        queries.append(data["question"])
 
+            # Test with more queries for robust measurement
+            test_queries = queries[:2000]
+            print(f"🧪 Testing with {len(test_queries)} queries")
+
+            # Test with complexity 64
+            complexity = 64
+            recall = evaluator.evaluate_recall_at_3(test_queries, complexity)
+            print(f"📈 Recall@3 at complexity {complexity}: {recall * 100:.1f}%")
+
+            evaluator.cleanup()
+            print("✅ Stage 2 completed!\n")
+
+        # Shared non-compact index path for Stage 3 and 4
+        non_compact_index_path = args.index.replace(".leann", "_noncompact.leann")
+        complexity = args.complexity
+
+        if args.stage == "3" or args.stage == "all":
+            # Stage 3: Binary search for 90% recall complexity (using non-compact index for speed)
+            print("🚀 Starting Stage 3: Binary search for 90% recall complexity")
+            print(
+                "💡 Creating non-compact index for fast binary search with recompute_embeddings=False"
+            )
+
+            # Create non-compact index for binary search (will be reused in Stage 4)
+            print("🏗️ Creating non-compact index for binary search...")
+            evaluator = FinanceBenchEvaluator(args.index)
+            evaluator.create_non_compact_index_for_comparison(non_compact_index_path)
+
+            # Use non-compact index for binary search
+            binary_search_evaluator = RecallEvaluator(non_compact_index_path, args.baseline_dir)
+
+            # Load queries for testing
+            print("📖 Loading FinanceBench dataset...")
+            queries = []
+            with open(args.dataset, encoding="utf-8") as f:
+                for line in f:
+                    if line.strip():
+                        data = json.loads(line)
+                        queries.append(data["question"])
+
+            # Use more queries for robust measurement
+            test_queries = queries[:200]
+            print(f"🧪 Testing with {len(test_queries)} queries")
+
+            # Binary search for 90% recall complexity (without recompute for speed)
+            target_recall = 0.9
+            min_complexity, max_complexity = 1, 32
+
+            print(f"🔍 Binary search for {target_recall * 100}% recall complexity...")
+            print(f"Search range: {min_complexity} to {max_complexity}")
+
+            best_complexity = None
+            best_recall = 0.0
+
+            while min_complexity <= max_complexity:
+                mid_complexity = (min_complexity + max_complexity) // 2
+
+                print(
+                    f"\n🧪 Testing complexity {mid_complexity} (no recompute, non-compact index)..."
+                )
+                # Use recompute_embeddings=False on non-compact index for fast binary search
+                recall = binary_search_evaluator.evaluate_recall_at_3(
+                    test_queries, mid_complexity, recompute_embeddings=False
+                )
+
+                print(
+                    f"  Complexity {mid_complexity}: Recall@3 = {recall:.3f} ({recall * 100:.1f}%)"
+                )
+
+                if recall >= target_recall:
+                    best_complexity = mid_complexity
+                    best_recall = recall
+                    max_complexity = mid_complexity - 1
+                    print("  ✅ Target reached! Searching for lower complexity...")
+                else:
+                    min_complexity = mid_complexity + 1
+                    print("  ❌ Below target. Searching for higher complexity...")
+
+            if best_complexity is not None:
+                print("\n🎯 Optimal complexity found!")
+                print(f"  Complexity: {best_complexity}")
+                print(f"  Recall@3: {best_recall:.3f} ({best_recall * 100:.1f}%)")
+
+                # Test a few complexities around the optimal one for verification
+                print("\n🔬 Verification test around optimal complexity:")
+                verification_complexities = [
+                    max(1, best_complexity - 2),
+                    max(1, best_complexity - 1),
+                    best_complexity,
+                    best_complexity + 1,
+                    best_complexity + 2,
+                ]
+
+                for complexity in verification_complexities:
+                    if complexity <= 512:  # reasonable upper bound
+                        recall = binary_search_evaluator.evaluate_recall_at_3(
+                            test_queries, complexity, recompute_embeddings=False
+                        )
+                        status = "✅" if recall >= target_recall else "❌"
+                        print(f"  {status} Complexity {complexity:3d}: {recall * 100:5.1f}%")
+
+                # Now test the optimal complexity with compact index and recompute for comparison
+                print(
+                    f"\n🔄 Testing optimal complexity {best_complexity} on compact index WITH recompute..."
+                )
+                compact_evaluator = RecallEvaluator(args.index, args.baseline_dir)
+                recall_with_recompute = compact_evaluator.evaluate_recall_at_3(
+                    test_queries[:10], best_complexity, recompute_embeddings=True
+                )
+                print(
+                    f"  ✅ Complexity {best_complexity} (compact index with recompute): {recall_with_recompute * 100:.1f}%"
+                )
+                complexity = best_complexity
+                print(
+                    f"  📊 Recall difference: {abs(best_recall - recall_with_recompute) * 100:.2f}%"
+                )
+                compact_evaluator.cleanup()
+            else:
+                print(f"\n❌ Could not find complexity achieving {target_recall * 100}% recall")
+                print("All tested complexities were below target.")
+
+            # Cleanup evaluators (keep non-compact index for Stage 4)
+            binary_search_evaluator.cleanup()
+            evaluator.cleanup()
+
+            print("✅ Stage 3 completed! Non-compact index saved for Stage 4.\n")
+
+        if args.stage == "4" or args.stage == "all":
+            # Stage 4: Comprehensive evaluation with dual index comparison
+            print("🚀 Starting Stage 4: Comprehensive evaluation with dual index comparison")
+
+            # Use FinanceBench evaluator for QA evaluation
+            evaluator = FinanceBenchEvaluator(args.index, args.openai_api_key)
+
+            print("📖 Loading FinanceBench dataset...")
+            data = evaluator.load_dataset(args.dataset)
+
+            # Step 1: Analyze current (compact) index
+            print("\n📏 Analyzing current index (compact, pruned)...")
+            compact_size_metrics = evaluator.analyze_index_sizes()
+            compact_size_metrics["index_type"] = "compact"
+
+            # Step 2: Use existing non-compact index or create if needed
+            from pathlib import Path
+
+            if Path(non_compact_index_path).exists():
+                print(
+                    f"\n📁 Using existing non-compact index from Stage 3: {non_compact_index_path}"
+                )
+                temp_evaluator = FinanceBenchEvaluator(non_compact_index_path)
+                non_compact_size_metrics = temp_evaluator.analyze_index_sizes()
+                non_compact_size_metrics["index_type"] = "non_compact"
+            else:
+                print("\n🏗️ Creating non-compact index (with embeddings) for comparison...")
+                non_compact_size_metrics = evaluator.create_non_compact_index_for_comparison(
+                    non_compact_index_path
+                )
+
+            # Step 3: Compare index sizes
+            print("\n📊 Index size comparison:")
+            print(
+                f"  Compact index (current): {compact_size_metrics['total_with_embeddings']:.1f} MB"
+            )
+            print(
+                f"  Non-compact index: {non_compact_size_metrics['total_with_embeddings']:.1f} MB"
+            )
+            size_increase = (
+                (
+                    non_compact_size_metrics["total_with_embeddings"]
+                    - compact_size_metrics["total_with_embeddings"]
+                )
+                / compact_size_metrics["total_with_embeddings"]
+                * 100
+            )
+            storage_saving = (
+                (
+                    non_compact_size_metrics["total_with_embeddings"]
+                    - compact_size_metrics["total_with_embeddings"]
+                )
+                / non_compact_size_metrics["total_with_embeddings"]
+                * 100
+            )
+            print(f"  Storage saving by compact: {storage_saving:.1f}%")
+
+            # Step 4: Performance comparison between the two indexes
+            if complexity is None:
+                raise ValueError("Complexity is required for performance comparison")
+
+            print("\n⚡ Performance comparison between indexes...")
+            performance_metrics = evaluator.compare_index_performance(
+                non_compact_index_path, args.index, data[:10], complexity=complexity
+            )
+
+            # Step 5: Timing breakdown evaluation WITH recompute (production mode)
+            test_samples = 20
+            print(f"\n🧪 Testing with first {test_samples} samples for timing analysis")
+            print(
+                "\n🔍🤖 Running timing breakdown evaluation (WITH recompute - production mode)..."
+            )
+            evaluation_start = time.time()
+            timing_metrics = evaluator.evaluate_timing_breakdown(data[:test_samples])
+            evaluation_time = time.time() - evaluation_start
+
+            # Combine all metrics
+            combined_metrics = {
+                **timing_metrics,
+                "total_evaluation_time": evaluation_time,
+                "current_index": compact_size_metrics,
+                "non_compact_index": non_compact_size_metrics,
+                "performance_comparison": performance_metrics,
+                "storage_saving_percent": storage_saving,
+            }
+
+            # Print comprehensive results
+            evaluator._print_results(combined_metrics)
+
+            # Save results if requested
+            if args.output:
+                print(f"\n💾 Saving results to {args.output}...")
+                with open(args.output, "w") as f:
+                    json.dump(combined_metrics, f, indent=2, default=str)
+                print(f"✅ Results saved to {args.output}")
+
+            evaluator.cleanup()
+            print("✅ Stage 4 completed!\n")
+
+        if args.stage == "all":
+            print("🎉 All evaluation stages completed successfully!")
+            print("\n📋 Summary:")
+            print("  Stage 2: ✅ Recall@3 evaluation completed")
+            print("  Stage 3: ✅ Optimal complexity found")
+            print("  Stage 4: ✅ Generation accuracy & timing evaluation completed")
+            print("\n🔧 Recommended next steps:")
+            print("  - Use optimal complexity for best speed/accuracy balance")
+            print("  - Review accuracy and timing breakdown for performance optimization")
+            print("  - Run full evaluation on complete dataset if needed")
+
+            # Clean up non-compact index after all stages complete
+            print("\n🧹 Cleaning up temporary non-compact index...")
+            from pathlib import Path
+
+            if Path(non_compact_index_path).exists():
+                temp_index_dir = Path(non_compact_index_path).parent
+                temp_index_name = Path(non_compact_index_path).name
+                for temp_file in temp_index_dir.glob(f"{temp_index_name}*"):
+                    temp_file.unlink()
+                print(f"✅ Cleaned up {non_compact_index_path}")
+            else:
+                print("📝 No temporary index to clean up")
     except KeyboardInterrupt:
         print("\n⚠️  Evaluation interrupted by user")
         exit(1)
     except Exception as e:
-        print(f"\n❌ Evaluation failed: {e}")
+        print(f"\n❌ Stage {args.stage} failed: {e}")
         exit(1)
 
 
diff --git a/benchmarks/financebench/setup_financebench.py b/benchmarks/financebench/setup_financebench.py
index e9f6028..36bd0da 100755
--- a/benchmarks/financebench/setup_financebench.py
+++ b/benchmarks/financebench/setup_financebench.py
@@ -5,6 +5,7 @@ Downloads all PDFs and builds full LEANN datastore
 """
 
 import argparse
+import os
 import re
 import time
 from concurrent.futures import ThreadPoolExecutor, as_completed
@@ -139,14 +140,15 @@ class FinanceBenchSetup:
 
         start_time = time.time()
 
-        # Initialize builder
+        # Initialize builder with standard compact configuration
         builder = LeannBuilder(
             backend_name=backend,
             embedding_model=embedding_model,
             embedding_mode="sentence-transformers",
             graph_degree=32,
             complexity=64,
-            is_recompute=False,  # Store embeddings for speed
+            is_recompute=True,  # Enable recompute (no stored embeddings)
+            is_compact=True,  # Enable compact storage (pruned)
             num_threads=4,
         )
 
@@ -185,6 +187,87 @@ class FinanceBenchSetup:
 
         return str(index_path)
 
+    def build_faiss_flat_baseline(self, index_path: str, output_dir: str = "baseline"):
+        """Build FAISS flat baseline using the same embeddings as LEANN index"""
+        print("🔨 Building FAISS Flat baseline...")
+
+        import os
+        import pickle
+
+        import numpy as np
+        from leann.api import compute_embeddings
+        from leann_backend_hnsw import faiss
+
+        os.makedirs(output_dir, exist_ok=True)
+        baseline_path = os.path.join(output_dir, "faiss_flat.index")
+        metadata_path = os.path.join(output_dir, "metadata.pkl")
+
+        if os.path.exists(baseline_path) and os.path.exists(metadata_path):
+            print(f"✅ Baseline already exists at {baseline_path}")
+            return baseline_path
+
+        # Read metadata from the built index
+        meta_path = f"{index_path}.meta.json"
+        with open(meta_path) as f:
+            import json
+
+            meta = json.loads(f.read())
+
+        embedding_model = meta["embedding_model"]
+        passage_source = meta["passage_sources"][0]
+        passage_file = passage_source["path"]
+
+        # Convert relative path to absolute
+        if not os.path.isabs(passage_file):
+            index_dir = os.path.dirname(index_path)
+            passage_file = os.path.join(index_dir, os.path.basename(passage_file))
+
+        print(f"📊 Loading passages from {passage_file}...")
+        print(f"🤖 Using embedding model: {embedding_model}")
+
+        # Load all passages for baseline
+        passages = []
+        passage_ids = []
+        with open(passage_file, encoding="utf-8") as f:
+            for line in f:
+                if line.strip():
+                    data = json.loads(line)
+                    passages.append(data["text"])
+                    passage_ids.append(data["id"])
+
+        print(f"📄 Loaded {len(passages)} passages")
+
+        # Compute embeddings using the same method as LEANN
+        print("🧮 Computing embeddings...")
+        embeddings = compute_embeddings(
+            passages,
+            embedding_model,
+            mode="sentence-transformers",
+            use_server=False,
+        )
+
+        print(f"📐 Embedding shape: {embeddings.shape}")
+
+        # Build FAISS flat index
+        print("🏗️  Building FAISS IndexFlatIP...")
+        dimension = embeddings.shape[1]
+        index = faiss.IndexFlatIP(dimension)
+
+        # Add embeddings to flat index
+        embeddings_f32 = embeddings.astype(np.float32)
+        index.add(embeddings_f32.shape[0], faiss.swig_ptr(embeddings_f32))
+
+        # Save index and metadata
+        faiss.write_index(index, baseline_path)
+        with open(metadata_path, "wb") as f:
+            pickle.dump(passage_ids, f)
+
+        print(f"✅ FAISS baseline saved to {baseline_path}")
+        print(f"✅ Metadata saved to {metadata_path}")
+        print(f"📊 Total vectors: {index.ntotal}")
+
+        return baseline_path
+
     def extract_pdf_text(self, pdf_path: Path) -> list[dict]:
         """Extract and chunk text from a PDF file"""
         chunks = []
@@ -300,6 +383,11 @@ def main():
     parser.add_argument("--max-workers", type=int, default=5, help="Parallel download workers")
     parser.add_argument("--skip-download", action="store_true", help="Skip PDF download")
     parser.add_argument("--skip-build", action="store_true", help="Skip index building")
+    parser.add_argument(
+        "--build-baseline-only",
+        action="store_true",
+        help="Only build FAISS baseline from existing index",
+    )
 
     args = parser.parse_args()
 
@@ -309,6 +397,24 @@ def main():
     setup = FinanceBenchSetup(args.data_dir)
 
     try:
+        if args.build_baseline_only:
+            # Only build baseline from existing index
+            index_path = setup.index_dir / f"financebench_full_{args.backend}"
+            index_file = f"{index_path}.index"
+            meta_file = f"{index_path}.leann.meta.json"
+
+            if not os.path.exists(index_file) or not os.path.exists(meta_file):
+                print("❌ Index files not found:")
+                print(f"   Index: {index_file}")
+                print(f"   Meta: {meta_file}")
+                print("💡 Run without --build-baseline-only to build the index first")
+                exit(1)
+
+            print(f"🔨 Building baseline from existing index: {index_path}")
+            baseline_path = setup.build_faiss_flat_baseline(str(index_path))
+            print(f"✅ Baseline built at {baseline_path}")
+            return
+
         # Step 1: Download dataset
         setup.download_dataset()
 
@@ -324,7 +430,12 @@ def main():
                 backend=args.backend, embedding_model=args.embedding_model
             )
 
-            # Step 4: Verify setup
+            # Step 4: Build FAISS flat baseline
+            print("\n🔨 Building FAISS flat baseline...")
+            baseline_path = setup.build_faiss_flat_baseline(index_path)
+            print(f"✅ Baseline built at {baseline_path}")
+
+            # Step 5: Verify setup
             setup.verify_setup(index_path)
         else:
             print("⏭️  Skipping index building")
diff --git a/benchmarks/financebench/verify_recall.py b/benchmarks/financebench/verify_recall.py
new file mode 100644
index 0000000..c4f77cb
--- /dev/null
+++ b/benchmarks/financebench/verify_recall.py
@@ -0,0 +1,214 @@
+#!/usr/bin/env python3
+# /// script
+# requires-python = ">=3.9"
+# dependencies = [
+#     "faiss-cpu",
+#     "numpy",
+#     "sentence-transformers",
+#     "torch",
+#     "tqdm",
+# ]
+# ///
+
+"""
+Independent recall verification script using standard FAISS.
+Creates two indexes (HNSW and Flat) and compares recall@3 at different complexities.
+"""
+
+import json
+import time
+from pathlib import Path
+
+import faiss
+import numpy as np
+from sentence_transformers import SentenceTransformer
+from tqdm import tqdm
+
+
+def compute_embeddings_direct(chunks: list[str], model_name: str) -> np.ndarray:
+    """
+    Direct embedding computation using sentence-transformers.
+    Copied logic to avoid dependency issues.
+    """
+    print(f"Loading model: {model_name}")
+    model = SentenceTransformer(model_name)
+
+    print(f"Computing embeddings for {len(chunks)} chunks...")
+    embeddings = model.encode(
+        chunks,
+        show_progress_bar=True,
+        batch_size=32,
+        convert_to_numpy=True,
+        normalize_embeddings=False,
+    )
+
+    return embeddings.astype(np.float32)
+
+
+def load_financebench_queries(dataset_path: str, max_queries: int = 200) -> list[str]:
+    """Load FinanceBench queries from dataset"""
+    queries = []
+    with open(dataset_path, encoding="utf-8") as f:
+        for line in f:
+            if line.strip():
+                data = json.loads(line)
+                queries.append(data["question"])
+                if len(queries) >= max_queries:
+                    break
+    return queries
+
+
+def load_passages_from_leann_index(index_path: str) -> tuple[list[str], list[str]]:
+    """Load passages from LEANN index structure"""
+    meta_path = f"{index_path}.meta.json"
+    with open(meta_path) as f:
+        meta = json.load(f)
+
+    passage_source = meta["passage_sources"][0]
+    passage_file = passage_source["path"]
+
+    # Convert relative path to absolute
+    if not Path(passage_file).is_absolute():
+        index_dir = Path(index_path).parent
+        passage_file = index_dir / Path(passage_file).name
+
+    print(f"Loading passages from {passage_file}")
+
+    passages = []
+    passage_ids = []
+    with open(passage_file, encoding="utf-8") as f:
+        for line in tqdm(f, desc="Loading passages"):
+            if line.strip():
+                data = json.loads(line)
+                passages.append(data["text"])
+                passage_ids.append(data["id"])
+
+    print(f"Loaded {len(passages)} passages")
+    return passages, passage_ids
+
+
+def build_faiss_indexes(embeddings: np.ndarray) -> tuple[faiss.Index, faiss.Index]:
+    """Build FAISS indexes: Flat (ground truth) and HNSW"""
+    dimension = embeddings.shape[1]
+
+    # Build Flat index (ground truth)
+    print("Building FAISS IndexFlatIP (ground truth)...")
+    flat_index = faiss.IndexFlatIP(dimension)
+    flat_index.add(embeddings)
+
+    # Build HNSW index
+    print("Building FAISS IndexHNSWFlat...")
+    M = 32  # Same as LEANN default
+    hnsw_index = faiss.IndexHNSWFlat(dimension, M, faiss.METRIC_INNER_PRODUCT)
+    hnsw_index.hnsw.efConstruction = 200  # Same as LEANN default
+    hnsw_index.add(embeddings)
+
+    print(f"Built indexes with {flat_index.ntotal} vectors, dimension {dimension}")
+    return flat_index, hnsw_index
+
+
+def evaluate_recall_at_k(
+    query_embeddings: np.ndarray,
+    flat_index: faiss.Index,
+    hnsw_index: faiss.Index,
+    passage_ids: list[str],
+    k: int = 3,
+    ef_search: int = 64,
+) -> float:
+    """Evaluate recall@k comparing HNSW vs Flat"""
+
+    # Set search parameters for HNSW
+    hnsw_index.hnsw.efSearch = ef_search
+
+    total_recall = 0.0
+    num_queries = query_embeddings.shape[0]
+
+    for i in range(num_queries):
+        query = query_embeddings[i : i + 1]  # Keep 2D shape
+
+        # Get ground truth from Flat index (standard FAISS API)
+        flat_distances, flat_indices = flat_index.search(query, k)
+        ground_truth_ids = {passage_ids[idx] for idx in flat_indices[0]}
+
+        # Get results from HNSW index (standard FAISS API)
+        hnsw_distances, hnsw_indices = hnsw_index.search(query, k)
+        hnsw_ids = {passage_ids[idx] for idx in hnsw_indices[0]}
+
+        # Calculate recall
+        intersection = ground_truth_ids.intersection(hnsw_ids)
+        recall = len(intersection) / k
+        total_recall += recall
+
+        if i < 3:  # Show first few examples
+            print(f"  Query {i + 1}: Recall@{k} = {recall:.3f}")
+            print(f"    Flat: {list(ground_truth_ids)}")
+            print(f"    HNSW: {list(hnsw_ids)}")
+            print(f"    Intersection: {list(intersection)}")
+
+    avg_recall = total_recall / num_queries
+    return avg_recall
+
+
+def main():
+    # Configuration
+    dataset_path = "data/financebench_merged.jsonl"
+    index_path = "data/index/financebench_full_hnsw.leann"
+    embedding_model = "sentence-transformers/all-mpnet-base-v2"
+
+    print("🔍 FAISS Recall Verification")
+    print("=" * 50)
+
+    # Check if files exist
+    if not Path(dataset_path).exists():
+        print(f"❌ Dataset not found: {dataset_path}")
+        return
+    if not Path(f"{index_path}.meta.json").exists():
+        print(f"❌ Index metadata not found: {index_path}.meta.json")
+        return
+
+    # Load data
+    print("📖 Loading FinanceBench queries...")
+    queries = load_financebench_queries(dataset_path, max_queries=50)
+    print(f"Loaded {len(queries)} queries")
+
+    print("📄 Loading passages from LEANN index...")
+    passages, passage_ids = load_passages_from_leann_index(index_path)
+
+    # Compute embeddings
+    print("🧮 Computing passage embeddings...")
+    passage_embeddings = compute_embeddings_direct(passages, embedding_model)
+
+    print("🧮 Computing query embeddings...")
+    query_embeddings = compute_embeddings_direct(queries, embedding_model)
+
+    # Build FAISS indexes
+    print("🏗️ Building FAISS indexes...")
+    flat_index, hnsw_index = build_faiss_indexes(passage_embeddings)
+
+    # Test different efSearch values (equivalent to LEANN complexity)
+    print("\n📊 Evaluating Recall@3 at different efSearch values...")
+    ef_search_values = [16, 32, 64, 128, 256]
+
+    for ef_search in ef_search_values:
+        print(f"\n🧪 Testing efSearch = {ef_search}")
+        start_time = time.time()
+
+        recall = evaluate_recall_at_k(
+            query_embeddings, flat_index, hnsw_index, passage_ids, k=3, ef_search=ef_search
+        )
+
+        elapsed = time.time() - start_time
+        print(
+            f"📈 efSearch {ef_search}: Recall@3 = {recall:.3f} ({recall * 100:.1f}%) in {elapsed:.2f}s"
+        )
+
+    print("\n✅ Verification completed!")
+    print("\n📋 Summary:")
+    print("  - Built independent FAISS Flat and HNSW indexes")
+    print("  - Compared recall@3 at different efSearch values")
+    print("  - Used same embedding model as LEANN")
+    print("  - This validates LEANN's recall measurements")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pyproject.toml b/pyproject.toml
index 3267332..961195e 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -46,6 +46,14 @@ dependencies = [
     "pathspec>=0.12.1",
     "nbconvert>=7.16.6",
     "gitignore-parser>=0.1.12",
+    # AST-aware code chunking dependencies
+    "astchunk>=0.1.0",
+    "tree-sitter>=0.20.0",
+    "tree-sitter-python>=0.20.0",
+    "tree-sitter-java>=0.20.0",
+    "tree-sitter-c-sharp>=0.20.0",
+    "tree-sitter-typescript>=0.20.0",
+    "faiss-cpu>=1.12.0",
 ]
 
 [project.optional-dependencies]
diff --git a/uv.lock b/uv.lock
index 535aa7c..438e7dd 100644
--- a/uv.lock
+++ b/uv.lock
@@ -1,5 +1,5 @@
 version = 1
-revision = 2
+revision = 3
 requires-python = ">=3.9"
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -198,6 +198,27 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/81/29/5ecc3a15d5a33e31b26c11426c45c501e439cb865d0bff96315d86443b78/appnope-0.1.4-py2.py3-none-any.whl", hash = "sha256:502575ee11cd7a28c0205f379b525beefebab9d161b7c964670864014ed7213c", size = 4321, upload-time = "2024-02-06T09:43:09.663Z" },
 ]
 
+[[package]]
+name = "astchunk"
+version = "0.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "numpy", version = "2.3.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "pyrsistent" },
+    { name = "tree-sitter", version = "0.23.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "tree-sitter", version = "0.25.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "tree-sitter-c-sharp" },
+    { name = "tree-sitter-java" },
+    { name = "tree-sitter-python" },
+    { name = "tree-sitter-typescript" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/db/2a/7a35e2fac7d550265ae2ee40651425083b37555f921d1a1b77c3f525e0df/astchunk-0.1.0.tar.gz", hash = "sha256:f4dff0ef8b3b3bcfeac363384db1e153f74d4c825dc2e35864abfab027713be4", size = 18093, upload-time = "2025-06-19T04:37:25.34Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/be/84/5433ab0e933b572750cb16fd7edf3d6c7902b069461a22ec670042752a4d/astchunk-0.1.0-py3-none-any.whl", hash = "sha256:33ada9fc3620807fdda5846fa1948af463f281a60e0d43d4f3782b6dbb416d24", size = 15396, upload-time = "2025-06-19T04:37:23.87Z" },
+]
+
 [[package]]
 name = "asttokens"
 version = "3.0.0"
@@ -1147,6 +1168,74 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/7b/8f/c4d9bafc34ad7ad5d8dc16dd1347ee0e507a52c3adb6bfa8887e1c6a26ba/executing-2.2.0-py2.py3-none-any.whl", hash = "sha256:11387150cad388d62750327a53d3339fad4888b39a6fe233c3afbb54ecffd3aa", size = 26702, upload-time = "2025-01-22T15:41:25.929Z" },
 ]
 
+[[package]]
+name = "faiss-cpu"
+version = "1.12.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "numpy", version = "2.3.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "packaging" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/7d/80/bb75a7ed6e824dea452a24d3434a72ed799324a688b10b047d441d270185/faiss_cpu-1.12.0.tar.gz", hash = "sha256:2f87cbcd603f3ed464ebceb857971fdebc318de938566c9ae2b82beda8e953c0", size = 69292, upload-time = "2025-08-13T06:07:26.553Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/bf/3b/42aa7332c2e432fc3af3a26cc49ca8a3ecd23d13bb790e61c1e54a4d16cb/faiss_cpu-1.12.0-cp310-cp310-macosx_13_0_x86_64.whl", hash = "sha256:be96f9290edd13d56fb3c69b8dd6be487552b4401f2e95b437cabf5309c424ad", size = 8006082, upload-time = "2025-08-13T06:05:33.131Z" },
+    { url = "https://files.pythonhosted.org/packages/00/ab/9959c2d9c3a511a5dbfa4e2e2a1d0bdcad5929d410b3abe87bbed74dcb9b/faiss_cpu-1.12.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:0834c547c39d5e5d0b769c90ac5d5ca42e00bcdbba491f3440d2d458058b19d6", size = 3360138, upload-time = "2025-08-13T06:05:35.003Z" },
+    { url = "https://files.pythonhosted.org/packages/80/b9/7456f89effe93b7693c7e39cd365065e27aa31794442510c44ad8cce6c4c/faiss_cpu-1.12.0-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a40830a16d8718b14a462e1a1efaa26660eb3bb8ada22e0712a6ac181092750e", size = 3825459, upload-time = "2025-08-13T06:05:36.546Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/0f/02d5d2ae8b53e5629cb03fbd871bbbfbbd647ffc3d09393b34f6347072d7/faiss_cpu-1.12.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d7f8732796e3f730556e99327861066ead0ae7e66b5cbf6c0f217be48074e41e", size = 31425823, upload-time = "2025-08-13T06:05:38.778Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/39/a9fcb0b82727ab2d5509caa7637e5d345c710502f68c7a7e90dd212654ad/faiss_cpu-1.12.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ded5063e13c3bb6b1b463827f838ae45a0aea4c9aeaf6c938e7e87f3f6ea4126", size = 9751939, upload-time = "2025-08-13T06:05:41.816Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/25/7efcb5856f9df4c003716687c4604bb5cfc44819539b79d60e302018962b/faiss_cpu-1.12.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:8e74e71249165757a12fb02feee67ea95df542bcafa21b449fbd2ed0c31b48b4", size = 24160951, upload-time = "2025-08-13T06:05:44.243Z" },
+    { url = "https://files.pythonhosted.org/packages/34/d4/1f1cc444708b426b42ec52f01c735d91cb9775fe55cf3d2c64b9a6fd8792/faiss_cpu-1.12.0-cp310-cp310-win_amd64.whl", hash = "sha256:d04d1cae2a9b66083cd8f48ff391731d81e0a1fdf67ab5c33ae10b3a22a0caae", size = 18169601, upload-time = "2025-08-13T06:05:46.558Z" },
+    { url = "https://files.pythonhosted.org/packages/87/ed/83fed257ea410c2e691374f04ac914d5f9414f04a9c7a266bdfbb999eb16/faiss_cpu-1.12.0-cp311-cp311-macosx_13_0_x86_64.whl", hash = "sha256:fbb63595c7ad43c0d9caaf4d554a38a30ea4edda5e7c3ed38845562776992ba9", size = 8006079, upload-time = "2025-08-13T06:05:48.932Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/07/80c248db87ef2e753ad390fca3b0d7dd6092079e904f35b248c7064e791e/faiss_cpu-1.12.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:83e74cbde6fa5caceec5bc103c82053d50fde163e3ceabaa58c91508e984142b", size = 3360138, upload-time = "2025-08-13T06:05:50.873Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/22/73bd9ed7b11cd14eb0da6e2f2eae763306abaad1b25a5808da8b1fc07665/faiss_cpu-1.12.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6155a5138604b702a32f8f0a63948a539eb7468898554a9911f9ab8c899284fb", size = 3825466, upload-time = "2025-08-13T06:05:52.311Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/7f/e1a21337b3cba24b953c760696e3b188a533d724440e050fd60a3c1aa919/faiss_cpu-1.12.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1bf4b5f0e9b6bb5a566b1a31e84a93b283f26c2b0155fb2eb5970c32a540a906", size = 31425626, upload-time = "2025-08-13T06:05:54.155Z" },
+    { url = "https://files.pythonhosted.org/packages/05/24/f352cf8400f414e6a31385ef12d43d11aac8beb11d573a2fd00ec44b8cb7/faiss_cpu-1.12.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:60a535b79d3d6225c7c21d7277fb0c6fde80c46a9c1e33632b1b293c1d177f30", size = 9751949, upload-time = "2025-08-13T06:05:56.369Z" },
+    { url = "https://files.pythonhosted.org/packages/05/50/a122e3076d7fd95cbe9a0cdf0fc796836f1e4fd399b418c6ba8533c75770/faiss_cpu-1.12.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0d1b243468a24564f85a41166f2ca4c92f8f6755da096ffbdcf551675ca739c5", size = 24161021, upload-time = "2025-08-13T06:05:58.776Z" },
+    { url = "https://files.pythonhosted.org/packages/72/9f/3344f6fe69f6fbfb19dec298b4dda3d47a87dc31e418911fdcc3a3ace013/faiss_cpu-1.12.0-cp311-cp311-win_amd64.whl", hash = "sha256:84510079a2efe954e6b89fe5e62f23a98c1ef999756565e056f95f835ff43c5e", size = 18169278, upload-time = "2025-08-13T06:06:01.44Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/b1/37d532292c1b3dab690636947a532d3797741b09f2dfb9cb558ffeaff34b/faiss_cpu-1.12.0-cp311-cp311-win_arm64.whl", hash = "sha256:2283f1014f7f86dd56b53bf0ea0d7f848eb4c9c6704b8f4f99a0af02e994e479", size = 8007093, upload-time = "2025-08-13T06:06:03.904Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/58/602ed184d35742eb240cbfea237bd214f2ae7f01cb369c39f4dff392f7c9/faiss_cpu-1.12.0-cp312-cp312-macosx_13_0_x86_64.whl", hash = "sha256:9b54990fcbcf90e37393909d4033520237194263c93ab6dbfae0616ef9af242b", size = 8034413, upload-time = "2025-08-13T06:06:05.564Z" },
+    { url = "https://files.pythonhosted.org/packages/83/d5/f84c3d0e022cdeb73ff8406a6834a7698829fa242eb8590ddf8a0b09357f/faiss_cpu-1.12.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:a5f5bca7e1a3e0a98480d1e2748fc86d12c28d506173e460e6746886ff0e08de", size = 3362034, upload-time = "2025-08-13T06:06:07.091Z" },
+    { url = "https://files.pythonhosted.org/packages/19/89/a4ba4d285ea4f9b0824bf31ebded3171da08bfcf5376f4771cc5481f72cd/faiss_cpu-1.12.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:016e391f49933875b8d60d47f282f2e93d8ea9f9ffbda82467aa771b11a237db", size = 3834319, upload-time = "2025-08-13T06:06:08.86Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/c9/be4e52fd96be601fefb313c26e1259ac2e6b556fb08cc392db641baba8c7/faiss_cpu-1.12.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c2e4963c7188f57cfba248f09ebd8a14c76b5ffb87382603ccd4576f2da39d74", size = 31421585, upload-time = "2025-08-13T06:06:10.643Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/aa/12c6723ce30df721a6bace21398559c0367c5418c04139babc2d26d8d158/faiss_cpu-1.12.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:88bfe134f8c7cd2dda7df34f2619448906624962c8207efdd6eb1647e2f5338b", size = 9762449, upload-time = "2025-08-13T06:06:13.373Z" },
+    { url = "https://files.pythonhosted.org/packages/67/15/ed2c9de47c3ebae980d6938f0ec12d739231438958bc5ab2d636b272d913/faiss_cpu-1.12.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:9243ee4c224a0d74419040503f22bf067462a040281bf6f3f107ab205c97d438", size = 24156525, upload-time = "2025-08-13T06:06:15.307Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/b8/6911de6b8fdcfa76144680c2195df6ce7e0cc920a8be8c5bbd2dfe5e3c37/faiss_cpu-1.12.0-cp312-cp312-win_amd64.whl", hash = "sha256:6b8012353d50d9bc81bcfe35b226d0e5bfad345fdebe0da31848395ebc83816d", size = 18169636, upload-time = "2025-08-13T06:06:17.613Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/69/d2b0f434b0ae35344280346b58d2b9a251609333424f3289c54506e60c51/faiss_cpu-1.12.0-cp312-cp312-win_arm64.whl", hash = "sha256:8b4f5b18cbe335322a51d2785bb044036609c35bfac5915bff95eadc10e89ef1", size = 8012423, upload-time = "2025-08-13T06:06:19.73Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/4e/6be5fbd2ceccd87b168c64edeefa469cd11f095bb63b16a61a29296b0fdb/faiss_cpu-1.12.0-cp313-cp313-macosx_13_0_x86_64.whl", hash = "sha256:c9c79b5f28dcf9b2e2557ce51b938b21b7a9d508e008dc1ffea7b8249e7bd443", size = 8034409, upload-time = "2025-08-13T06:06:22.519Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/f0/658012a91a690d82f3587fd8e56ea1d9b9698c31970929a9dba17edd211e/faiss_cpu-1.12.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:0db6485bc9f32b69aaccf9ad520782371a79904dcfe20b6da5cbfd61a712e85f", size = 3362034, upload-time = "2025-08-13T06:06:24.052Z" },
+    { url = "https://files.pythonhosted.org/packages/81/8b/9b355309d448e1a737fac31d45e9b2484ffb0f04f10fba3b544efe6661e4/faiss_cpu-1.12.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f6db5532831791d7bac089fc580e741e99869122946bb6a5f120016c83b95d10", size = 3834324, upload-time = "2025-08-13T06:06:25.506Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/31/d229f6cdb9cbe03020499d69c4b431b705aa19a55aa0fe698c98022b2fef/faiss_cpu-1.12.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4d57ed7aac048b18809af70350c31acc0fb9f00e6c03b6ed1651fd58b174882d", size = 31421590, upload-time = "2025-08-13T06:06:27.601Z" },
+    { url = "https://files.pythonhosted.org/packages/26/19/80289ba008f14c95fbb6e94617ea9884e421ca745864fe6b8b90e1c3fc94/faiss_cpu-1.12.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:26c29290e7d1c5938e5886594dc0a2272b30728351ca5f855d4ae30704d5a6cc", size = 9762452, upload-time = "2025-08-13T06:06:30.237Z" },
+    { url = "https://files.pythonhosted.org/packages/af/e7/6cc03ead5e19275e34992419e2b7d107d0295390ccf589636ff26adb41e2/faiss_cpu-1.12.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:9b43d0c295e93a8e5f1dd30325caaf34d4ecb51f1e3d461c7b0e71bff3a8944b", size = 24156530, upload-time = "2025-08-13T06:06:32.23Z" },
+    { url = "https://files.pythonhosted.org/packages/34/90/438865fe737d65e7348680dadf3b2983bdcef7e5b7e852000e74c50a9933/faiss_cpu-1.12.0-cp313-cp313-win_amd64.whl", hash = "sha256:a7c6156f1309bb969480280906e8865c3c4378eebb0f840c55c924bf06efd8d3", size = 18169604, upload-time = "2025-08-13T06:06:34.884Z" },
+    { url = "https://files.pythonhosted.org/packages/76/69/40a1d8d781a70d33c57ef1b4b777486761dd1c502a86d27e90ef6aa8a9f9/faiss_cpu-1.12.0-cp313-cp313-win_arm64.whl", hash = "sha256:0b5fac98a350774a98b904f7a7c6689eb5cf0a593d63c552e705a80c55636d15", size = 8012523, upload-time = "2025-08-13T06:06:37.24Z" },
+    { url = "https://files.pythonhosted.org/packages/12/35/01a4a7c179d67bee0d8a027b95c3eae19cb354ae69ef2bc50ac3b93bc853/faiss_cpu-1.12.0-cp314-cp314-macosx_13_0_x86_64.whl", hash = "sha256:ff7db774968210d08cd0331287f3f66a6ffef955a7aa9a7fcd3eb4432a4ce5f5", size = 8036142, upload-time = "2025-08-13T06:06:38.894Z" },
+    { url = "https://files.pythonhosted.org/packages/08/23/bac2859490096608c9d527f3041b44c2e43f8df0d4aadd53a4cc5ce678ac/faiss_cpu-1.12.0-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:220b5bb5439c64e417b35f9ade4c7dc3bf7df683d6123901ba84d6d764ecd486", size = 3363747, upload-time = "2025-08-13T06:06:40.73Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/1d/e18023e1f43a18ec593adcd69d356f1fa94bde20344e38334d5985e5c5cc/faiss_cpu-1.12.0-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:693d0bf16f79e8d16a1baaeda459f3375f37da0354e97dc032806b48a2a54151", size = 3835232, upload-time = "2025-08-13T06:06:42.172Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/2b/1c1fea423d3f550f44c5ec3f14d8400919b49c285c3bd146687c63e40186/faiss_cpu-1.12.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:bcc6587dee21e17430fb49ddc5200625d6f5e1de2bdf436f14827bad4ca78d19", size = 31432677, upload-time = "2025-08-13T06:06:44.348Z" },
+    { url = "https://files.pythonhosted.org/packages/de/d2/3483e92a02f30e2d8491a256f470f54b7f5483266dfe09126d28741d31ec/faiss_cpu-1.12.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:b80e5965f001822cc99ec65c715169af1b70bdae72eccd573520a2dec485b3ee", size = 9765504, upload-time = "2025-08-13T06:06:46.567Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/2f/d97792211a9bd84b8d6b1dcaa1dcd69ac11e026c6ef19c641b6a87e31025/faiss_cpu-1.12.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:98279f1b4876ef9902695a329b81a99002782ab6e26def472022009df6f1ac68", size = 24169930, upload-time = "2025-08-13T06:06:48.916Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/b8/b707ca4d88af472509a053c39d3cced53efd19d096b8dff2fadc18c4b82d/faiss_cpu-1.12.0-cp314-cp314-win_amd64.whl", hash = "sha256:11670337f9f5ee9ff3490e30683eea80add060c300cf6f6cb0e8faf3155fd20e", size = 18475400, upload-time = "2025-08-13T06:06:51.233Z" },
+    { url = "https://files.pythonhosted.org/packages/77/11/42e41ddebde4dfe77e36e92d0110b4f733c8640883abffde54f802482deb/faiss_cpu-1.12.0-cp314-cp314-win_arm64.whl", hash = "sha256:7ac1c8b53609b5c722ab60f1749260a7cb3c72fdfb720a0e3033067e73591da5", size = 8281229, upload-time = "2025-08-13T06:06:53.735Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/9a/8ae5bbeabe70eb673c37fc7c77e2e476746331afb6654b2df97d8b6d380d/faiss_cpu-1.12.0-cp314-cp314t-macosx_13_0_x86_64.whl", hash = "sha256:110b21b7bb4c93c4f1a5eb2ffb8ef99dcdb4725f8ab2e5cd161324e4d981f204", size = 8087247, upload-time = "2025-08-13T06:06:55.407Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/df/b3d79098860b67b126da351788c04ac243c29718dadc4a678a6f5e7209c0/faiss_cpu-1.12.0-cp314-cp314t-macosx_14_0_arm64.whl", hash = "sha256:82eb5515ce72be9a43f4cf74447a0d090e014231981df91aff7251204b506fbf", size = 3411043, upload-time = "2025-08-13T06:06:56.983Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/2f/b1a2a03dd3cce22ff9fc434aa3c7390125087260c1d1349311da36eaa432/faiss_cpu-1.12.0-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:754eef89cdf2b35643df6b0923a5a098bdfecf63b5f4bd86c385042ee511b287", size = 3801789, upload-time = "2025-08-13T06:06:58.688Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/a8/16ad0c6a966e93d04bfd5248d2be1d8b5849842b0e2611c5ecd26fcaf036/faiss_cpu-1.12.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7285c71c8f5e9c58b55175f5f74c78c518c52c421a88a430263f34e3e31f719c", size = 31231388, upload-time = "2025-08-13T06:07:00.55Z" },
+    { url = "https://files.pythonhosted.org/packages/62/a1/9c16eca0b8f8b13c32c47a5e4ff7a4bc0ca3e7d263140312088811230871/faiss_cpu-1.12.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:84a50d7a2f711f79cc8b65aa28956dba6435e47b71a38b2daea44c94c9b8e458", size = 9737605, upload-time = "2025-08-13T06:07:03.018Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/4a/2c2d615078c9d816a836fb893aaef551ad152f2eb00bc258698273c240c0/faiss_cpu-1.12.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:7f3e0a14e4edec6a3959a9f51afccb89e863138f184ff2cc24c13f9ad788740b", size = 23922880, upload-time = "2025-08-13T06:07:05.099Z" },
+    { url = "https://files.pythonhosted.org/packages/30/aa/99b8402a4dac678794f13f8f4f29d666c2ef0a91594418147f47034ebc81/faiss_cpu-1.12.0-cp314-cp314t-win_amd64.whl", hash = "sha256:8b3239cc371df6826ac43c62ac04eec7cc497bedb43f681fcd8ea494f520ddbb", size = 18750661, upload-time = "2025-08-13T06:07:07.551Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/a2/b546e9a20ba157eb2fbe141289f1752f157ee6d932899f4853df4ded6d4b/faiss_cpu-1.12.0-cp314-cp314t-win_arm64.whl", hash = "sha256:58b23456db725ee1bd605a6135d2ef55b2ac3e0b6fe873fd99a909e8ef4bd0ff", size = 8302032, upload-time = "2025-08-13T06:07:09.602Z" },
+    { url = "https://files.pythonhosted.org/packages/95/5a/1ec3ee947526d95cb5656ef3343d7022f92704eee45f38249ee540378d44/faiss_cpu-1.12.0-cp39-cp39-macosx_13_0_x86_64.whl", hash = "sha256:564c1b707fa9e68e4551cd78cbc6dc89810d740cccfc3076c34a0ba6627119ef", size = 8006142, upload-time = "2025-08-13T06:07:11.476Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/94/9f3a0e09e5f8c360dcea076be3c9a30c55e7fb23cb2741c12f70921bc061/faiss_cpu-1.12.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:6a80f9223f1f5eea42a853355520ff0efd38d00923dd9e581195b9ecf1095f0e", size = 3359833, upload-time = "2025-08-13T06:07:13.387Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/65/0ba324613bd234dc48d51da9e0124b7e97de242251135207f368b473062e/faiss_cpu-1.12.0-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4b9d38691646da1ba12e3c0be8eb910ac52e2bcb0e553b54590c1c7aded9d5a3", size = 3824212, upload-time = "2025-08-13T06:07:15.315Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/03/17777b9cb92db034e49804522ec176bbd4898cdca0c7c1ebf483d5817bee/faiss_cpu-1.12.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:433ddfd09f5ef3cd2593b03a39f37e6957b25a91792b17132d3414c99d5ba5dd", size = 31425806, upload-time = "2025-08-13T06:07:17.51Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/c9/87a8f360436378238074e9980d876d8e5e1dea86eeaa66fab8e4acb62d79/faiss_cpu-1.12.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:fc98084076d0ceffcc3e306fd77ee22f91cc6143953e471c31c84cf2ddf68bb2", size = 9751334, upload-time = "2025-08-13T06:07:19.777Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/89/f9145aee30ab0728abb16ee4374469eb50a65ea004c26321097c70818d4a/faiss_cpu-1.12.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c5867ee3132cdc9f279951b8514ab51fd121acfbea120c9948b06bd1a1e43053", size = 24160811, upload-time = "2025-08-13T06:07:22.123Z" },
+    { url = "https://files.pythonhosted.org/packages/62/cf/41c9912052afce0c3de2be92639ba609bd80f7cabde7e1abd0f80664010b/faiss_cpu-1.12.0-cp39-cp39-win_amd64.whl", hash = "sha256:9bcc6e9eb111fc5b06f1623f4073b5acdae25a5baee1eea32b347a2d810deb35", size = 18169958, upload-time = "2025-08-13T06:07:24.365Z" },
+]
+
 [[package]]
 name = "fastjsonschema"
 version = "2.21.1"
@@ -2058,7 +2147,7 @@ wheels = [
 
 [[package]]
 name = "leann-backend-diskann"
-version = "0.3.0"
+version = "0.3.2"
 source = { editable = "packages/leann-backend-diskann" }
 dependencies = [
     { name = "leann-core" },
@@ -2070,14 +2159,14 @@ dependencies = [
 
 [package.metadata]
 requires-dist = [
-    { name = "leann-core", specifier = "==0.3.0" },
+    { name = "leann-core", specifier = "==0.3.2" },
     { name = "numpy" },
     { name = "protobuf", specifier = ">=3.19.0" },
 ]
 
 [[package]]
 name = "leann-backend-hnsw"
-version = "0.3.0"
+version = "0.3.2"
 source = { editable = "packages/leann-backend-hnsw" }
 dependencies = [
     { name = "leann-core" },
@@ -2090,7 +2179,7 @@ dependencies = [
 
 [package.metadata]
 requires-dist = [
-    { name = "leann-core", specifier = "==0.3.0" },
+    { name = "leann-core", specifier = "==0.3.2" },
     { name = "msgpack", specifier = ">=1.0.0" },
     { name = "numpy" },
     { name = "pyzmq", specifier = ">=23.0.0" },
@@ -2098,7 +2187,7 @@ requires-dist = [
 
 [[package]]
 name = "leann-core"
-version = "0.3.0"
+version = "0.3.2"
 source = { editable = "packages/leann-core" }
 dependencies = [
     { name = "accelerate" },
@@ -2164,10 +2253,12 @@ name = "leann-workspace"
 version = "0.1.0"
 source = { editable = "." }
 dependencies = [
+    { name = "astchunk" },
     { name = "boto3" },
     { name = "colorama" },
     { name = "datasets" },
     { name = "evaluate" },
+    { name = "faiss-cpu" },
     { name = "gitignore-parser" },
     { name = "ipykernel" },
     { name = "leann-backend-hnsw" },
@@ -2198,6 +2289,12 @@ dependencies = [
     { name = "sglang" },
     { name = "torch" },
     { name = "tqdm" },
+    { name = "tree-sitter", version = "0.23.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "tree-sitter", version = "0.25.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "tree-sitter-c-sharp" },
+    { name = "tree-sitter-java" },
+    { name = "tree-sitter-python" },
+    { name = "tree-sitter-typescript" },
     { name = "typer" },
 ]
 
@@ -2231,12 +2328,14 @@ test = [
 
 [package.metadata]
 requires-dist = [
+    { name = "astchunk", specifier = ">=0.1.0" },
     { name = "beautifulsoup4", marker = "extra == 'documents'", specifier = ">=4.13.0" },
     { name = "black", marker = "extra == 'dev'", specifier = ">=23.0" },
     { name = "boto3" },
     { name = "colorama" },
     { name = "datasets", specifier = ">=2.15.0" },
     { name = "evaluate" },
+    { name = "faiss-cpu", specifier = ">=1.12.0" },
     { name = "gitignore-parser", specifier = ">=0.1.12" },
     { name = "huggingface-hub", marker = "extra == 'dev'", specifier = ">=0.20.0" },
     { name = "ipykernel", specifier = "==6.29.5" },
@@ -2280,6 +2379,11 @@ requires-dist = [
     { name = "sglang" },
     { name = "torch" },
     { name = "tqdm" },
+    { name = "tree-sitter", specifier = ">=0.20.0" },
+    { name = "tree-sitter-c-sharp", specifier = ">=0.20.0" },
+    { name = "tree-sitter-java", specifier = ">=0.20.0" },
+    { name = "tree-sitter-python", specifier = ">=0.20.0" },
+    { name = "tree-sitter-typescript", specifier = ">=0.20.0" },
     { name = "typer", specifier = ">=0.12.3" },
 ]
 provides-extras = ["dev", "test", "diskann", "documents"]
@@ -4441,6 +4545,39 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/be/7a/097801205b991bc3115e8af1edb850d30aeaf0118520b016354cf5ccd3f6/pypdfium2-4.30.0-py3-none-win_arm64.whl", hash = "sha256:119b2969a6d6b1e8d55e99caaf05290294f2d0fe49c12a3f17102d01c441bd29", size = 2752118, upload-time = "2024-05-09T18:33:15.489Z" },
 ]
 
+[[package]]
+name = "pyrsistent"
+version = "0.20.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ce/3a/5031723c09068e9c8c2f0bc25c3a9245f2b1d1aea8396c787a408f2b95ca/pyrsistent-0.20.0.tar.gz", hash = "sha256:4c48f78f62ab596c679086084d0dd13254ae4f3d6c72a83ffdf5ebdef8f265a4", size = 103642, upload-time = "2023-10-25T21:06:56.342Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c7/19/c343b14061907b629b765444b6436b160e2bd4184d17d4804bbe6381f6be/pyrsistent-0.20.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:8c3aba3e01235221e5b229a6c05f585f344734bd1ad42a8ac51493d74722bbce", size = 83416, upload-time = "2023-10-25T21:06:04.579Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/4f/8342079ea331031ef9ed57edd312a9ad283bcc8adfaf268931ae356a09a6/pyrsistent-0.20.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c1beb78af5423b879edaf23c5591ff292cf7c33979734c99aa66d5914ead880f", size = 118021, upload-time = "2023-10-25T21:06:06.953Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/b7/64a125c488243965b7c5118352e47c6f89df95b4ac306d31cee409153d57/pyrsistent-0.20.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:21cc459636983764e692b9eba7144cdd54fdec23ccdb1e8ba392a63666c60c34", size = 117747, upload-time = "2023-10-25T21:06:08.5Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/a5/43c67bd5f80df9e7583042398d12113263ec57f27c0607abe9d78395d18f/pyrsistent-0.20.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f5ac696f02b3fc01a710427585c855f65cd9c640e14f52abe52020722bb4906b", size = 114524, upload-time = "2023-10-25T21:06:10.728Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/98/b382a87e89ca839106d874f7bf78d226b3eedb26735eb6f751f1a3375f21/pyrsistent-0.20.0-cp310-cp310-win32.whl", hash = "sha256:0724c506cd8b63c69c7f883cc233aac948c1ea946ea95996ad8b1380c25e1d3f", size = 60780, upload-time = "2023-10-25T21:06:12.14Z" },
+    { url = "https://files.pythonhosted.org/packages/37/8a/23e2193f7adea6901262e3cf39c7fe18ac0c446176c0ff0e19aeb2e9681e/pyrsistent-0.20.0-cp310-cp310-win_amd64.whl", hash = "sha256:8441cf9616d642c475684d6cf2520dd24812e996ba9af15e606df5f6fd9d04a7", size = 63310, upload-time = "2023-10-25T21:06:13.598Z" },
+    { url = "https://files.pythonhosted.org/packages/df/63/7544dc7d0953294882a5c587fb1b10a26e0c23d9b92281a14c2514bac1f7/pyrsistent-0.20.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:0f3b1bcaa1f0629c978b355a7c37acd58907390149b7311b5db1b37648eb6958", size = 83481, upload-time = "2023-10-25T21:06:15.238Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/a0/49249bc14d71b1bf2ffe89703acfa86f2017c25cfdabcaea532b8c8a5810/pyrsistent-0.20.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5cdd7ef1ea7a491ae70d826b6cc64868de09a1d5ff9ef8d574250d0940e275b8", size = 120222, upload-time = "2023-10-25T21:06:17.144Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/94/9808e8c9271424120289b9028a657da336ad7e43da0647f62e4f6011d19b/pyrsistent-0.20.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cae40a9e3ce178415040a0383f00e8d68b569e97f31928a3a8ad37e3fde6df6a", size = 120002, upload-time = "2023-10-25T21:06:18.727Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/f6/9ecfb78b2fc8e2540546db0fe19df1fae0f56664a5958c21ff8861b0f8da/pyrsistent-0.20.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6288b3fa6622ad8a91e6eb759cfc48ff3089e7c17fb1d4c59a919769314af224", size = 116850, upload-time = "2023-10-25T21:06:20.424Z" },
+    { url = "https://files.pythonhosted.org/packages/83/c8/e6d28bc27a0719f8eaae660357df9757d6e9ca9be2691595721de9e8adfc/pyrsistent-0.20.0-cp311-cp311-win32.whl", hash = "sha256:7d29c23bdf6e5438c755b941cef867ec2a4a172ceb9f50553b6ed70d50dfd656", size = 60775, upload-time = "2023-10-25T21:06:21.815Z" },
+    { url = "https://files.pythonhosted.org/packages/98/87/c6ef52ff30388f357922d08de012abdd3dc61e09311d88967bdae23ab657/pyrsistent-0.20.0-cp311-cp311-win_amd64.whl", hash = "sha256:59a89bccd615551391f3237e00006a26bcf98a4d18623a19909a2c48b8e986ee", size = 63306, upload-time = "2023-10-25T21:06:22.874Z" },
+    { url = "https://files.pythonhosted.org/packages/15/ee/ff2ed52032ac1ce2e7ba19e79bd5b05d152ebfb77956cf08fcd6e8d760ea/pyrsistent-0.20.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:09848306523a3aba463c4b49493a760e7a6ca52e4826aa100ee99d8d39b7ad1e", size = 83537, upload-time = "2023-10-25T21:06:24.17Z" },
+    { url = "https://files.pythonhosted.org/packages/80/f1/338d0050b24c3132bcfc79b68c3a5f54bce3d213ecef74d37e988b971d8a/pyrsistent-0.20.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a14798c3005ec892bbada26485c2eea3b54109cb2533713e355c806891f63c5e", size = 122615, upload-time = "2023-10-25T21:06:25.815Z" },
+    { url = "https://files.pythonhosted.org/packages/07/3a/e56d6431b713518094fae6ff833a04a6f49ad0fbe25fb7c0dc7408e19d20/pyrsistent-0.20.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b14decb628fac50db5e02ee5a35a9c0772d20277824cfe845c8a8b717c15daa3", size = 122335, upload-time = "2023-10-25T21:06:28.631Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/bb/5f40a4d5e985a43b43f607250e766cdec28904682c3505eb0bd343a4b7db/pyrsistent-0.20.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2e2c116cc804d9b09ce9814d17df5edf1df0c624aba3b43bc1ad90411487036d", size = 118510, upload-time = "2023-10-25T21:06:30.718Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/13/e6a22f40f5800af116c02c28e29f15c06aa41cb2036f6a64ab124647f28b/pyrsistent-0.20.0-cp312-cp312-win32.whl", hash = "sha256:e78d0c7c1e99a4a45c99143900ea0546025e41bb59ebc10182e947cf1ece9174", size = 60865, upload-time = "2023-10-25T21:06:32.742Z" },
+    { url = "https://files.pythonhosted.org/packages/75/ef/2fa3b55023ec07c22682c957808f9a41836da4cd006b5f55ec76bf0fbfa6/pyrsistent-0.20.0-cp312-cp312-win_amd64.whl", hash = "sha256:4021a7f963d88ccd15b523787d18ed5e5269ce57aa4037146a2377ff607ae87d", size = 63239, upload-time = "2023-10-25T21:06:34.035Z" },
+    { url = "https://files.pythonhosted.org/packages/18/0c/289126299fcebf54fd01d385fb5176c328fef2c4233139c23dd48346e992/pyrsistent-0.20.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:ca52d1ceae015859d16aded12584c59eb3825f7b50c6cfd621d4231a6cc624ce", size = 83379, upload-time = "2023-10-25T21:06:45.585Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/45/62639d53ac09eaafc00f2e5845565e70d3eddb2d296337a77637186ca03e/pyrsistent-0.20.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b318ca24db0f0518630e8b6f3831e9cba78f099ed5c1d65ffe3e023003043ba0", size = 117740, upload-time = "2023-10-25T21:06:46.918Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/12/24b9a6ef7b991b6722756e0aa169a39463af2b8ed0fb526f0a00aae34ea4/pyrsistent-0.20.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fed2c3216a605dc9a6ea50c7e84c82906e3684c4e80d2908208f662a6cbf9022", size = 117457, upload-time = "2023-10-25T21:06:48.911Z" },
+    { url = "https://files.pythonhosted.org/packages/19/3c/ab06510f86bc0934b77ade41948924ff1f33dcd3433f32feca2028218837/pyrsistent-0.20.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2e14c95c16211d166f59c6611533d0dacce2e25de0f76e4c140fde250997b3ca", size = 114280, upload-time = "2023-10-25T21:06:50.503Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/b1/1275bbfb929854d20e72aa2bbfb50ea3b1d7d41a95848b353691875e2817/pyrsistent-0.20.0-cp39-cp39-win32.whl", hash = "sha256:f058a615031eea4ef94ead6456f5ec2026c19fb5bd6bfe86e9665c4158cf802f", size = 60764, upload-time = "2023-10-25T21:06:52.093Z" },
+    { url = "https://files.pythonhosted.org/packages/28/77/0d7af973c0e3b1b83d8b45943601f77f85b943007e3a4d8744f7102c652b/pyrsistent-0.20.0-cp39-cp39-win_amd64.whl", hash = "sha256:58b8f6366e152092194ae68fefe18b9f0b4f89227dfd86a07770c3d86097aebf", size = 63289, upload-time = "2023-10-25T21:06:53.221Z" },
+    { url = "https://files.pythonhosted.org/packages/23/88/0acd180010aaed4987c85700b7cc17f9505f3edb4e5873e4dc67f613e338/pyrsistent-0.20.0-py3-none-any.whl", hash = "sha256:c55acc4733aad6560a7f5f818466631f07efc001fd023f34a6c203f8b6df0f0b", size = 58106, upload-time = "2023-10-25T21:06:54.387Z" },
+]
+
 [[package]]
 name = "pytest"
 version = "8.4.1"
@@ -5827,6 +5964,165 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/cc/34/4d82dc596764de9d14285f8ed53b50896bf05fbbcd71a82c6d174b3ab8c7/transformers-4.54.0-py3-none-any.whl", hash = "sha256:c96e607f848625965b76c677b2c2576f2c7b7097c1c5292b281919d90675a25e", size = 11176597, upload-time = "2025-07-25T18:58:17.677Z" },
 ]
 
+[[package]]
+name = "tree-sitter"
+version = "0.23.2"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/0f/50/fd5fafa42b884f741b28d9e6fd366c3f34e15d2ed3aa9633b34e388379e2/tree-sitter-0.23.2.tar.gz", hash = "sha256:66bae8dd47f1fed7bdef816115146d3a41c39b5c482d7bad36d9ba1def088450", size = 166800, upload-time = "2024-10-24T15:31:02.238Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/91/04/2068a7b725265ecfcbf63ecdae038f1d4124ebccd55b8a7ce145b70e2b6a/tree_sitter-0.23.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:3a937f5d8727bc1c74c4bf2a9d1c25ace049e8628273016ad0d45914ae904e10", size = 139289, upload-time = "2024-10-24T15:29:59.27Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/07/a5b943121f674fe1ac77694a698e71ce95353830c1f3f4ce45da7ef3e406/tree_sitter-0.23.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:2c7eae7fe2af215645a38660d2d57d257a4c461fe3ec827cca99a79478284e80", size = 132379, upload-time = "2024-10-24T15:30:01.437Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/96/fcc72c33d464a2d722db1e95b74a53ced771a47b3cfde60aced29764a783/tree_sitter-0.23.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3a71d607595270b6870eaf778a1032d146b2aa79bfcfa60f57a82a7b7584a4c7", size = 552884, upload-time = "2024-10-24T15:30:02.672Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/af/b0e787a52767155b4643a55d6de03c1e4ae77abb61e1dc1629ad983e0a40/tree_sitter-0.23.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6fe9b9ea7a0aa23b52fd97354da95d1b2580065bc12a4ac868f9164a127211d6", size = 566561, upload-time = "2024-10-24T15:30:04.073Z" },
+    { url = "https://files.pythonhosted.org/packages/65/fd/05e966b5317b1c6679c071c5b0203f28af9d26c9363700cb9682e1bcf343/tree_sitter-0.23.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d74d00a8021719eae14d10d1b1e28649e15d8b958c01c2b2c3dad7a2ebc4dbae", size = 558273, upload-time = "2024-10-24T15:30:06.177Z" },
+    { url = "https://files.pythonhosted.org/packages/60/bc/19145efdf3f47711aa3f1bf06f0b50593f97f1108550d38694841fd97b7c/tree_sitter-0.23.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:6de18d8d8a7f67ab71f472d1fcb01cc506e080cbb5e13d52929e4b6fdce6bbee", size = 569176, upload-time = "2024-10-24T15:30:07.902Z" },
+    { url = "https://files.pythonhosted.org/packages/32/08/3553d8e488ae9284a0762effafb7d2639a306e184963b7f99853923084d6/tree_sitter-0.23.2-cp310-cp310-win_amd64.whl", hash = "sha256:12b60dca70d2282af942b650a6d781be487485454668c7c956338a367b98cdee", size = 117902, upload-time = "2024-10-24T15:30:09.675Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/39/836fa485e985c33e8aa1cc3abbf7a84be1c2c382e69547a765631fdd7ce3/tree_sitter-0.23.2-cp310-cp310-win_arm64.whl", hash = "sha256:3346a4dd0447a42aabb863443b0fd8c92b909baf40ed2344fae4b94b625d5955", size = 102644, upload-time = "2024-10-24T15:30:11.484Z" },
+    { url = "https://files.pythonhosted.org/packages/55/8d/2d4fb04408772be0919441d66f700673ce7cb76b9ab6682e226d740fb88d/tree_sitter-0.23.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:91fda41d4f8824335cc43c64e2c37d8089c8c563bd3900a512d2852d075af719", size = 139142, upload-time = "2024-10-24T15:30:12.627Z" },
+    { url = "https://files.pythonhosted.org/packages/32/52/b8a44bfff7b0203256e5dbc8d3a372ee8896128b8ed7d3a89e1ef17b2065/tree_sitter-0.23.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:92b2b489d5ce54b41f94c6f23fbaf592bd6e84dc2877048fd1cb060480fa53f7", size = 132198, upload-time = "2024-10-24T15:30:13.893Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/54/746f2ee5acf6191a4a0be7f5843329f0d713bfe5196f5fc6fe2ea69cb44c/tree_sitter-0.23.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:64859bd4aa1567d0d6016a811b2b49c59d4a4427d096e3d8c84b2521455f62b7", size = 554303, upload-time = "2024-10-24T15:30:15.334Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/5a/3169d9933be813776a9b4b3f2e671d3d50fa27e589dee5578f6ecef7ff6d/tree_sitter-0.23.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:614590611636044e071d3a0b748046d52676dbda3bc9fa431216231e11dd98f7", size = 567626, upload-time = "2024-10-24T15:30:17.12Z" },
+    { url = "https://files.pythonhosted.org/packages/32/0d/23f363b3b0bc3fa0e7a4a294bf119957ac1ab02737d57815e1e8b7b3e196/tree_sitter-0.23.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:08466953c78ae57be61057188fb88c89791b0a562856010228e0ccf60e2ac453", size = 559803, upload-time = "2024-10-24T15:30:18.921Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/b3/1ffba0f17a7ff2c9114d91a1ecc15e0748f217817797564d31fbb61d7458/tree_sitter-0.23.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:8a33f03a562de91f7fd05eefcedd8994a06cd44c62f7aabace811ad82bc11cbd", size = 570987, upload-time = "2024-10-24T15:30:21.116Z" },
+    { url = "https://files.pythonhosted.org/packages/59/4b/085bcb8a11ea18003aacc4dbc91c301d1536c5e2deedb95393e8ef26f1f7/tree_sitter-0.23.2-cp311-cp311-win_amd64.whl", hash = "sha256:03b70296b569ef64f7b92b42ca5da9bf86d81bee2afd480bea35092687f51dae", size = 117771, upload-time = "2024-10-24T15:30:22.38Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/e5/90adc4081f49ccb6bea89a800dc9b0dcc5b6953b0da423e8eff28f63fddf/tree_sitter-0.23.2-cp311-cp311-win_arm64.whl", hash = "sha256:7cb4bb953ea7c0b50eeafc4454783e030357179d2a93c3dd5ebed2da5588ddd0", size = 102555, upload-time = "2024-10-24T15:30:23.534Z" },
+    { url = "https://files.pythonhosted.org/packages/07/a7/57e0fe87b49a78c670a7b4483f70e44c000c65c29b138001096b22e7dd87/tree_sitter-0.23.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:a014498b6a9e6003fae8c6eb72f5927d62da9dcb72b28b3ce8cd15c6ff6a6572", size = 139259, upload-time = "2024-10-24T15:30:24.941Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/b9/bc8513d818ffb54993a017a36c8739300bc5739a13677acf90b54995e7db/tree_sitter-0.23.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:04f8699b131d4bcbe3805c37e4ef3d159ee9a82a0e700587625623999ba0ea53", size = 131951, upload-time = "2024-10-24T15:30:26.176Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/6a/eab01bb6b1ce3c9acf16d72922ffc29a904af485eb3e60baf3a3e04edd30/tree_sitter-0.23.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4471577df285059c71686ecb208bc50fb472099b38dcc8e849b0e86652891e87", size = 557952, upload-time = "2024-10-24T15:30:27.389Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/95/f2f73332623cf63200d57800f85273170bc5f99d28ea3f234afd5b0048df/tree_sitter-0.23.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f342c925290dd4e20ecd5787ef7ae8749981597ab364783a1eb73173efe65226", size = 571199, upload-time = "2024-10-24T15:30:28.879Z" },
+    { url = "https://files.pythonhosted.org/packages/04/ac/bd6e6cfdd0421156e86f5c93848629af1c7323083077e1a95b27d32d5811/tree_sitter-0.23.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a4e9e53d07dd076bede72e4f7d3a0173d7b9ad6576572dd86da008a740a9bb22", size = 562129, upload-time = "2024-10-24T15:30:30.199Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/bd/8a9edcbcf8a76b0bf58e3b927ed291e3598e063d56667367762833cc8709/tree_sitter-0.23.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:8caebe65bc358759dac2500d8f8feed3aed939c4ade9a684a1783fe07bc7d5db", size = 574307, upload-time = "2024-10-24T15:30:32.085Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/c2/3fb2c6c0ae2f59a7411dc6d3e7945e3cb6f34c8552688708acc8b2b13f83/tree_sitter-0.23.2-cp312-cp312-win_amd64.whl", hash = "sha256:fc5a72eb50d43485000dbbb309acb350467b7467e66dc747c6bb82ce63041582", size = 117858, upload-time = "2024-10-24T15:30:33.353Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/18/4ca2c0f4a0c802ebcb3a92264cc436f1d54b394fa24dfa76bf57cdeaca9e/tree_sitter-0.23.2-cp312-cp312-win_arm64.whl", hash = "sha256:a0320eb6c7993359c5f7b371d22719ccd273f440d41cf1bd65dac5e9587f2046", size = 102496, upload-time = "2024-10-24T15:30:34.782Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/c6/4ead9ce3113a7c27f37a2bdef163c09757efbaa85adbdfe7b3fbf0317c57/tree_sitter-0.23.2-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:eff630dddee7ba05accb439b17e559e15ce13f057297007c246237ceb6306332", size = 139266, upload-time = "2024-10-24T15:30:35.946Z" },
+    { url = "https://files.pythonhosted.org/packages/76/c9/b4197c5b0c1d6ba648202a547846ac910a53163b69a459504b2aa6cdb76e/tree_sitter-0.23.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:4780ba8f3894f2dea869fad2995c2aceab3fd5ab9e6a27c45475d2acd7f7e84e", size = 131959, upload-time = "2024-10-24T15:30:37.646Z" },
+    { url = "https://files.pythonhosted.org/packages/99/94/0f7c5580d2adff3b57d36f1998725b0caf6cf1af50ceafc00c6cdbc2fef6/tree_sitter-0.23.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f0b609460b8e3e256361fb12e94fae5b728cb835b16f0f9d590b5aadbf9d109b", size = 557582, upload-time = "2024-10-24T15:30:39.019Z" },
+    { url = "https://files.pythonhosted.org/packages/97/8a/f73ff06959d43fd47fc283cbcc4d8efa6550b2cc431d852b184504992447/tree_sitter-0.23.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:78d070d8eaeaeb36cf535f55e5578fddbfc3bf53c1980f58bf1a99d57466b3b5", size = 570891, upload-time = "2024-10-24T15:30:40.432Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/86/bbda5ad09b88051ff7bf3275622a2f79bc4f728b4c283ff8b93b8fcdf36d/tree_sitter-0.23.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:878580b2ad5054c410ba3418edca4d34c81cc26706114d8f5b5541688bc2d785", size = 562343, upload-time = "2024-10-24T15:30:43.045Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/55/b404fa49cb5c2926ad6fe1cac033dd486ef69f1afeb7828452d21e1e05c1/tree_sitter-0.23.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:29224bdc2a3b9af535b7725e249d3ee291b2e90708e82832e73acc175e40dc48", size = 574407, upload-time = "2024-10-24T15:30:45.018Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/c8/eea2104443ab973091107ef3e730683bd8e6cb51dd025cef853d3fff9dae/tree_sitter-0.23.2-cp313-cp313-win_amd64.whl", hash = "sha256:c58d89348162fbc3aea1fe6511a66ee189fc0e4e4bbe937026f29e4ecef17763", size = 117854, upload-time = "2024-10-24T15:30:47.817Z" },
+    { url = "https://files.pythonhosted.org/packages/89/4d/1728d9ce32a1d851081911b7e47830f5e740431f2bb920f54bb8c26175bc/tree_sitter-0.23.2-cp313-cp313-win_arm64.whl", hash = "sha256:0ff2037be5edab7801de3f6a721b9cf010853f612e2008ee454e0e0badb225a6", size = 102492, upload-time = "2024-10-24T15:30:48.892Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/ab/b39173a47d498cc6276e303c865f4a222134ceae890bd3c1b29427489805/tree_sitter-0.23.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a5db8e585205faef8bf219da77d8993e2ef04d08eda2e3c8ad7e4df8297ee344", size = 139550, upload-time = "2024-10-24T15:30:50.516Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/34/fa8f5b862dd7a6014fd5578810178e8f7601830cabb6d65d2aba050c2df1/tree_sitter-0.23.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:9dbd110a30cf28be5da734ae4cd0e9031768228dbf6a79f2973962aa51de4ec7", size = 132686, upload-time = "2024-10-24T15:30:51.779Z" },
+    { url = "https://files.pythonhosted.org/packages/98/b9/ccdddf35705fc23395caa71557f767e0753d38afe4b5bb99efddbf62bb22/tree_sitter-0.23.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:569514b9a996a0fd458b3a891c46ca125298be0c03cf82f2b6f0c13d5d8f25dc", size = 554958, upload-time = "2024-10-24T15:30:53.327Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/ba/20ae9079bdfc5cfac28b39d945a6c354c8e1385e73aec8142db6c53b635c/tree_sitter-0.23.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a357ed98a74e47787b812df99a74a2c35c0fe11e55c2095cc01d1cad144ef552", size = 568162, upload-time = "2024-10-24T15:30:54.667Z" },
+    { url = "https://files.pythonhosted.org/packages/40/00/b16bf6cf88c47c1b6c8e1cce1eb9e90badb5db9e5252ae0970d858d02592/tree_sitter-0.23.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:c2dfb8e8f760f4cc67888d03ef9e2dbd3353245f67f5efba375c2a14d944ac0e", size = 560278, upload-time = "2024-10-24T15:30:56.49Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/8f/27ab9b96cc0261af78b080ec8a9846a38e216360ec38774ea27eba35bd3c/tree_sitter-0.23.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:3ead958df87a21d706903987e665e9e0e5df7b2c5021ff69ea349826840adc6a", size = 571255, upload-time = "2024-10-24T15:30:58.254Z" },
+    { url = "https://files.pythonhosted.org/packages/44/e0/95a3d66a7e5bb229574484ab10c6dc99d1c7a32972b890d194076e30dc4f/tree_sitter-0.23.2-cp39-cp39-win_amd64.whl", hash = "sha256:611cae16be332213c0e6ece72c0bfca202e30ff320a8b309b1526c6cb79ee4ba", size = 118232, upload-time = "2024-10-24T15:30:59.965Z" },
+    { url = "https://files.pythonhosted.org/packages/10/b5/9eaf794fc71490573ab14a366affca415bc1ddbf86a14d78e54583db4254/tree_sitter-0.23.2-cp39-cp39-win_arm64.whl", hash = "sha256:b848e0fdd522fbb8888cdb4f4d93f8fad97ae10d70c122fb922e51363c7febcd", size = 102787, upload-time = "2024-10-24T15:31:01.084Z" },
+]
+
+[[package]]
+name = "tree-sitter"
+version = "0.25.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/89/2b/02a642e67605b9dd59986b00d13a076044dede04025a243f0592ac79d68c/tree-sitter-0.25.1.tar.gz", hash = "sha256:cd761ad0e4d1fc88a4b1b8083bae06d4f973acf6f5f29bbf13ea9609c1dec9c1", size = 177874, upload-time = "2025-08-05T17:14:34.193Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e0/6c/6160ca15926d11a6957d8bee887f477f3c1d9bc5272c863affc0b50b9cff/tree_sitter-0.25.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:a15d62ffdb095d509bda8c140c1ddd0cc80f0c67f92b87fcc96cd242dc0c71ea", size = 146692, upload-time = "2025-08-05T17:13:54.559Z" },
+    { url = "https://files.pythonhosted.org/packages/81/4a/e5eb39fe73a514a13bf94acee97925de296d673dace00557763cbbdc938f/tree_sitter-0.25.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:1d938f0a1ffad1206a1a569b0501345eeca81cae0a4487bb485e53768b02f24e", size = 141015, upload-time = "2025-08-05T17:13:55.807Z" },
+    { url = "https://files.pythonhosted.org/packages/63/22/c8e3ba245e5cdb8c951482028a7ee99d141302047b708dc9d670f0fafd85/tree_sitter-0.25.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ba8cea296de5dcb384b9a15cf526985ac8339c81da51c7e29a251d82071f5ee9", size = 599462, upload-time = "2025-08-05T17:13:56.984Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/91/c866c3d278ee86354fd81fd055b5d835c510b0e9af07e1cf7e48e2f946b0/tree_sitter-0.25.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:387fd2bd8657d69e877618dc199c18e2d6fe073b8f5c59e23435f3baee4ee10a", size = 627062, upload-time = "2025-08-05T17:13:58.363Z" },
+    { url = "https://files.pythonhosted.org/packages/90/96/ac010f72778dae60381ab5fcca9651ac72647d582db0b027ca6c56116920/tree_sitter-0.25.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:afa49e51f82b58ae2c1291d6b79ca31e0fb36c04bd9a20d89007472edfb70136", size = 623788, upload-time = "2025-08-05T17:13:59.431Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/29/190bdfd54a564a2e43a702884ad5679f4578c481a46161f9f335dd390a70/tree_sitter-0.25.1-cp310-cp310-win_amd64.whl", hash = "sha256:77be45f666adf284914510794b41100decccd71dba88010c03dc2bb0d653acec", size = 127253, upload-time = "2025-08-05T17:14:00.446Z" },
+    { url = "https://files.pythonhosted.org/packages/da/60/7daca5ccf65fb204c9f2cc2907db6aeaf1cb42aa605427580c17a38a53b3/tree_sitter-0.25.1-cp310-cp310-win_arm64.whl", hash = "sha256:72badac2de4e81ae0df5efe14ec5003bd4df3e48e7cf84dbd9df3a54599ba371", size = 113930, upload-time = "2025-08-05T17:14:01.623Z" },
+    { url = "https://files.pythonhosted.org/packages/17/dc/0dabb75d249108fb9062d6e9e791e4ad8e9ae5c095e06dd8af770bc07902/tree_sitter-0.25.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:33a8fbaeb2b5049cf5318306ab8b16ab365828b2b21ee13678c29e0726a1d27a", size = 146696, upload-time = "2025-08-05T17:14:02.408Z" },
+    { url = "https://files.pythonhosted.org/packages/da/d0/b7305a05d65dbcfce7a97a93252bf7384f09800866e9de55a625c76e0257/tree_sitter-0.25.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:797bbbc686d8d3722d25ee0108ad979bda6ad3e1025859ce2ee290e517816bd4", size = 141014, upload-time = "2025-08-05T17:14:03.58Z" },
+    { url = "https://files.pythonhosted.org/packages/84/d0/d0d8bd13c44ef6379499712a3f5e3930e7db11e5c8eb2af8655e288597a3/tree_sitter-0.25.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:629fc2ae3f5954b0f6a7b42ee3fcd8f34b68ea161e9f02fa5bf709cbbac996d3", size = 604339, upload-time = "2025-08-05T17:14:04.722Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/13/22869a6da25ffe2dfff922712605e72a9c3481109a93f4218bea1bc65f35/tree_sitter-0.25.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4257018c42a33a7935a5150d678aac05c6594347d6a6e6dbdf7e2ef4ae985213", size = 631593, upload-time = "2025-08-05T17:14:06.043Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/0c/f4590fc08422768fc57456a85c932888a02e7a13540574859308611be1cf/tree_sitter-0.25.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4027854c9feee2a3bb99642145ba04ce95d75bd17e292911c93a488cb28d0a04", size = 629265, upload-time = "2025-08-05T17:14:07.045Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/a8/ee9305ce9a7417715cbf038fdcc4fdb6042e30065c9837bdcf36be440388/tree_sitter-0.25.1-cp311-cp311-win_amd64.whl", hash = "sha256:183faaedcee5f0a3ba39257fa81749709d5eb7cf92c2c050b36ff38468d1774c", size = 127210, upload-time = "2025-08-05T17:14:08.331Z" },
+    { url = "https://files.pythonhosted.org/packages/48/64/6a39882f534373873ef3dba8a1a8f47dc3bfb39ee63784eac2e789b404c4/tree_sitter-0.25.1-cp311-cp311-win_arm64.whl", hash = "sha256:6a3800235535a2532ce392ed0d8e6f698ee010e73805bdeac2f249da8246bab6", size = 113928, upload-time = "2025-08-05T17:14:09.376Z" },
+    { url = "https://files.pythonhosted.org/packages/45/79/6dea0c098879d99f41ba919da1ea46e614fb4bf9c4d591450061aeec6fcb/tree_sitter-0.25.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:9362a202144075b54f7c9f07e0b0e44a61eed7ee19e140c506b9e64c1d21ed58", size = 146928, upload-time = "2025-08-05T17:14:10.522Z" },
+    { url = "https://files.pythonhosted.org/packages/15/30/8002f4e76c7834a6101895ff7524ea29ab4f1f1da1270260ef52e2319372/tree_sitter-0.25.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:593f22529f34dd04de02f56ea6d7c2c8ec99dfab25b58be893247c1090dedd60", size = 140802, upload-time = "2025-08-05T17:14:11.38Z" },
+    { url = "https://files.pythonhosted.org/packages/38/ec/d297ad9d4a4b26f551a5ca49afe48fdbcb20f058c2eff8d8463ad6c0eed1/tree_sitter-0.25.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ebb6849f76e1cbfa223303fa680da533d452e378d5fe372598e4752838ca7929", size = 606762, upload-time = "2025-08-05T17:14:12.264Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/1c/05a623cfb420b10d5f782d4ec064cf00fbfa9c21b8526ca4fd042f80acff/tree_sitter-0.25.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:034d4544bb0f82e449033d76dd083b131c3f9ecb5e37d3475f80ae55e8f382bd", size = 634632, upload-time = "2025-08-05T17:14:13.21Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/e0/f05fd5a2331c16d428efb8eef32dfb80dc6565438146e34e9a235ecd7925/tree_sitter-0.25.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:46a9b721560070f2f980105266e28a17d3149485582cdba14d66dca14692e932", size = 630756, upload-time = "2025-08-05T17:14:14.673Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/fc/79f3c5d53d1721b95ab6cda0368192a4f1d367e3a5ff7ac21d77e9841782/tree_sitter-0.25.1-cp312-cp312-win_amd64.whl", hash = "sha256:9a5c522b1350a626dc1cbc5dc203133caeaa114d3f65e400445e8b02f18b343b", size = 127157, upload-time = "2025-08-05T17:14:15.59Z" },
+    { url = "https://files.pythonhosted.org/packages/24/b7/07c4e3f71af0096db6c2ecd83e7d61584e3891c79cb39b208082312d1d60/tree_sitter-0.25.1-cp312-cp312-win_arm64.whl", hash = "sha256:43e7b8e83f9fc29ca62e7d2aa8c38e3fa806ff3fc65e0d501d18588dc1509888", size = 113910, upload-time = "2025-08-05T17:14:16.385Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/d3/bfb08aab9c7daed2715f303cc017329e3512bb77678cc28829681decadd2/tree_sitter-0.25.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ae1eebc175e6a50b38b0e0385cdc26e92ac0bff9b32ee1c0619bbbf6829d57ea", size = 146920, upload-time = "2025-08-05T17:14:17.483Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/36/7f897c50489c38665255579646fca8191e1b9e5a29ac9cf11022e42e1e2b/tree_sitter-0.25.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:9e0ae03c4f132f1bffb2bc40b1bb28742785507da693ab04da8531fe534ada9c", size = 140782, upload-time = "2025-08-05T17:14:18.594Z" },
+    { url = "https://files.pythonhosted.org/packages/16/e6/85012113899296b8e0789ae94f562d3971d7d3df989e8bec6128749394e1/tree_sitter-0.25.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:acf571758be0a71046a61a0936cb815f15b13e0ae7ec6d08398e4aa1560b371d", size = 607590, upload-time = "2025-08-05T17:14:19.782Z" },
+    { url = "https://files.pythonhosted.org/packages/49/93/605b08dc4cf76d08cfacebc30a88467c6526ea5c94592c25240518e38b71/tree_sitter-0.25.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:632910847e3f8ae35841f92cba88a9a1b8bc56ecc1514a5affebf7951fa0fc0a", size = 635553, upload-time = "2025-08-05T17:14:21.107Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/27/123667f756bb32168507c940db9040104c606fbb0214397d3c20cf985073/tree_sitter-0.25.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a99ecef7771afb118b2a8435c8ba67ea7a085c60d5d33dc0a4794ed882e5f7df", size = 630844, upload-time = "2025-08-05T17:14:22.078Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/53/180b0ed74153a3c9a23967f54774d5930c2e0b67671ae4ca0d4d35ba18ac/tree_sitter-0.25.1-cp313-cp313-win_amd64.whl", hash = "sha256:c1d6393454d1f9d4195c74e40a487640cd4390cd4aee90837485f932a1a0f40c", size = 127159, upload-time = "2025-08-05T17:14:23.061Z" },
+    { url = "https://files.pythonhosted.org/packages/32/fb/b8b7b5122ac4a80cd689a5023f2416910e10f9534ace1cdf0020a315d40d/tree_sitter-0.25.1-cp313-cp313-win_arm64.whl", hash = "sha256:c1d2dbf7d12426b71ff49739f599c355f4de338a5c0ab994de2a1d290f6e0b20", size = 113920, upload-time = "2025-08-05T17:14:23.879Z" },
+    { url = "https://files.pythonhosted.org/packages/70/8c/cb851da552baf4215baf96443e5e9e39095083a95bc05c4444e640fe0fe8/tree_sitter-0.25.1-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:32cee52264d9ecf98885fcac0185ac63e16251b31dd8b4a3b8d8071173405f8f", size = 146775, upload-time = "2025-08-05T17:14:25.064Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/59/002c89df1e8f1664b82023e5d0c06de97fff5c2a2e33dce1a241c8909758/tree_sitter-0.25.1-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:ae024d8ccfef51e61c44a81af7a48670601430701c24f450bea10f4b4effd8d1", size = 140787, upload-time = "2025-08-05T17:14:25.914Z" },
+    { url = "https://files.pythonhosted.org/packages/39/48/c9e6deb88f3c7f16963ef205e5b8e3ea7f5effd048b4515d09738c7b032b/tree_sitter-0.25.1-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d025c56c393cea660df9ef33ca60329952a1f8ee6212d21b2b390dfec08a3874", size = 609173, upload-time = "2025-08-05T17:14:26.817Z" },
+    { url = "https://files.pythonhosted.org/packages/53/a8/b782576d7ea081a87285d974005155da03b6d0c66283fe1e3a5e0dd4bd98/tree_sitter-0.25.1-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:044aa23ea14f337809821bea7467f33f4c6d351739dca76ba0cbe4d0154d8662", size = 635994, upload-time = "2025-08-05T17:14:28.343Z" },
+    { url = "https://files.pythonhosted.org/packages/70/0a/c5b6c9cdb7bd4bf0c3d2bd494fcf356acc53f8e63007dc2a836d95bbe964/tree_sitter-0.25.1-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:1863d96704eb002df4ad3b738294ae8bd5dcf8cefb715da18bff6cb2d33d978e", size = 630944, upload-time = "2025-08-05T17:14:31.123Z" },
+    { url = "https://files.pythonhosted.org/packages/12/2a/d0b097157c2d487f5e6293dae2c106ec9ede792a6bb780249e81432e754d/tree_sitter-0.25.1-cp314-cp314-win_amd64.whl", hash = "sha256:a40a481e28e1afdbc455932d61e49ffd4163aafa83f4a3deb717524a7786197e", size = 130831, upload-time = "2025-08-05T17:14:32.458Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/33/3591e7b22dd49f46ae4fdee1db316ecefd0486cae880c5b497a55f0ccb24/tree_sitter-0.25.1-cp314-cp314-win_arm64.whl", hash = "sha256:f7b68f584336b39b2deab9896b629dddc3c784170733d3409f01fe825e9c04eb", size = 117376, upload-time = "2025-08-05T17:14:33.283Z" },
+]
+
+[[package]]
+name = "tree-sitter-c-sharp"
+version = "0.23.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/22/85/a61c782afbb706a47d990eaee6977e7c2bd013771c5bf5c81c617684f286/tree_sitter_c_sharp-0.23.1.tar.gz", hash = "sha256:322e2cfd3a547a840375276b2aea3335fa6458aeac082f6c60fec3f745c967eb", size = 1317728, upload-time = "2024-11-11T05:25:32.535Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/58/04/f6c2df4c53a588ccd88d50851155945cff8cd887bd70c175e00aaade7edf/tree_sitter_c_sharp-0.23.1-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:2b612a6e5bd17bb7fa2aab4bb6fc1fba45c94f09cb034ab332e45603b86e32fd", size = 372235, upload-time = "2024-11-11T05:25:19.424Z" },
+    { url = "https://files.pythonhosted.org/packages/99/10/1aa9486f1e28fc22810fa92cbdc54e1051e7f5536a5e5b5e9695f609b31e/tree_sitter_c_sharp-0.23.1-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:1a8b98f62bc53efcd4d971151950c9b9cd5cbe3bacdb0cd69fdccac63350d83e", size = 419046, upload-time = "2024-11-11T05:25:20.679Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/21/13df29f8fcb9ba9f209b7b413a4764b673dfd58989a0dd67e9c7e19e9c2e/tree_sitter_c_sharp-0.23.1-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:986e93d845a438ec3c4416401aa98e6a6f6631d644bbbc2e43fcb915c51d255d", size = 415999, upload-time = "2024-11-11T05:25:22.359Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/72/fc6846795bcdae2f8aa94cc8b1d1af33d634e08be63e294ff0d6794b1efc/tree_sitter_c_sharp-0.23.1-cp39-abi3-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a8024e466b2f5611c6dc90321f232d8584893c7fb88b75e4a831992f877616d2", size = 402830, upload-time = "2024-11-11T05:25:24.198Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/3a/b6028c5890ce6653807d5fa88c72232c027c6ceb480dbeb3b186d60e5971/tree_sitter_c_sharp-0.23.1-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:7f9bf876866835492281d336b9e1f9626ab668737f74e914c31d285261507da7", size = 397880, upload-time = "2024-11-11T05:25:25.937Z" },
+    { url = "https://files.pythonhosted.org/packages/47/d2/4facaa34b40f8104d8751746d0e1cd2ddf0beb9f1404b736b97f372bd1f3/tree_sitter_c_sharp-0.23.1-cp39-abi3-win_amd64.whl", hash = "sha256:ae9a9e859e8f44e2b07578d44f9a220d3fa25b688966708af6aa55d42abeebb3", size = 377562, upload-time = "2024-11-11T05:25:27.539Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/88/3cf6bd9959d94d1fec1e6a9c530c5f08ff4115a474f62aedb5fedb0f7241/tree_sitter_c_sharp-0.23.1-cp39-abi3-win_arm64.whl", hash = "sha256:c81548347a93347be4f48cb63ec7d60ef4b0efa91313330e69641e49aa5a08c5", size = 375157, upload-time = "2024-11-11T05:25:30.839Z" },
+]
+
+[[package]]
+name = "tree-sitter-java"
+version = "0.23.5"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/fa/dc/eb9c8f96304e5d8ae1663126d89967a622a80937ad2909903569ccb7ec8f/tree_sitter_java-0.23.5.tar.gz", hash = "sha256:f5cd57b8f1270a7f0438878750d02ccc79421d45cca65ff284f1527e9ef02e38", size = 138121, upload-time = "2024-12-21T18:24:26.936Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/67/21/b3399780b440e1567a11d384d0ebb1aea9b642d0d98becf30fa55c0e3a3b/tree_sitter_java-0.23.5-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:355ce0308672d6f7013ec913dee4a0613666f4cda9044a7824240d17f38209df", size = 58926, upload-time = "2024-12-21T18:24:12.53Z" },
+    { url = "https://files.pythonhosted.org/packages/57/ef/6406b444e2a93bc72a04e802f4107e9ecf04b8de4a5528830726d210599c/tree_sitter_java-0.23.5-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:24acd59c4720dedad80d548fe4237e43ef2b7a4e94c8549b0ca6e4c4d7bf6e69", size = 62288, upload-time = "2024-12-21T18:24:14.634Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/6c/74b1c150d4f69c291ab0b78d5dd1b59712559bbe7e7daf6d8466d483463f/tree_sitter_java-0.23.5-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9401e7271f0b333df39fc8a8336a0caf1b891d9a2b89ddee99fae66b794fc5b7", size = 85533, upload-time = "2024-12-21T18:24:16.695Z" },
+    { url = "https://files.pythonhosted.org/packages/29/09/e0d08f5c212062fd046db35c1015a2621c2631bc8b4aae5740d7adb276ad/tree_sitter_java-0.23.5-cp39-abi3-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:370b204b9500b847f6d0c5ad584045831cee69e9a3e4d878535d39e4a7e4c4f1", size = 84033, upload-time = "2024-12-21T18:24:18.758Z" },
+    { url = "https://files.pythonhosted.org/packages/43/56/7d06b23ddd09bde816a131aa504ee11a1bbe87c6b62ab9b2ed23849a3382/tree_sitter_java-0.23.5-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:aae84449e330363b55b14a2af0585e4e0dae75eb64ea509b7e5b0e1de536846a", size = 82564, upload-time = "2024-12-21T18:24:20.493Z" },
+    { url = "https://files.pythonhosted.org/packages/da/d6/0528c7e1e88a18221dbd8ccee3825bf274b1fa300f745fd74eb343878043/tree_sitter_java-0.23.5-cp39-abi3-win_amd64.whl", hash = "sha256:1ee45e790f8d31d416bc84a09dac2e2c6bc343e89b8a2e1d550513498eedfde7", size = 60650, upload-time = "2024-12-21T18:24:22.902Z" },
+    { url = "https://files.pythonhosted.org/packages/72/57/5bab54d23179350356515526fff3cc0f3ac23bfbc1a1d518a15978d4880e/tree_sitter_java-0.23.5-cp39-abi3-win_arm64.whl", hash = "sha256:402efe136104c5603b429dc26c7e75ae14faaca54cfd319ecc41c8f2534750f4", size = 59059, upload-time = "2024-12-21T18:24:24.934Z" },
+]
+
+[[package]]
+name = "tree-sitter-python"
+version = "0.23.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/1c/30/6766433b31be476fda6569a3a374c2220e45ffee0bff75460038a57bf23b/tree_sitter_python-0.23.6.tar.gz", hash = "sha256:354bfa0a2f9217431764a631516f85173e9711af2c13dbd796a8815acfe505d9", size = 155868, upload-time = "2024-12-22T23:09:55.918Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ab/67/577a02acae5f776007c924ca86ef14c19c12e71de0aa9d2a036f3c248e7b/tree_sitter_python-0.23.6-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:28fbec8f74eeb2b30292d97715e60fac9ccf8a8091ce19b9d93e9b580ed280fb", size = 74361, upload-time = "2024-12-22T23:09:42.37Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/a6/194b3625a7245c532ad418130d63077ce6cd241152524152f533e4d6edb0/tree_sitter_python-0.23.6-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:680b710051b144fedf61c95197db0094f2245e82551bf7f0c501356333571f7a", size = 76436, upload-time = "2024-12-22T23:09:43.566Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/62/1da112689d6d282920e62c40e67ab39ea56463b0e7167bfc5e81818a770e/tree_sitter_python-0.23.6-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8a9dcef55507b6567207e8ee0a6b053d0688019b47ff7f26edc1764b7f4dc0a4", size = 112060, upload-time = "2024-12-22T23:09:44.721Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/62/c9358584c96e38318d69b6704653684fd8467601f7b74e88aa44f4e6903f/tree_sitter_python-0.23.6-cp39-abi3-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:29dacdc0cd2f64e55e61d96c6906533ebb2791972bec988450c46cce60092f5d", size = 112338, upload-time = "2024-12-22T23:09:48.323Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/58/c5e61add45e34fb8ecbf057c500bae9d96ed7c9ca36edb7985da8ae45526/tree_sitter_python-0.23.6-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:7e048733c36f564b379831689006801feb267d8194f9e793fbb395ef1723335d", size = 109382, upload-time = "2024-12-22T23:09:49.49Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/f3/9b30893cae9b3811fe652dc6f90aaadfda12ae0b2757f5722fc7266f423c/tree_sitter_python-0.23.6-cp39-abi3-win_amd64.whl", hash = "sha256:a24027248399fb41594b696f929f9956828ae7cc85596d9f775e6c239cd0c2be", size = 75904, upload-time = "2024-12-22T23:09:51.597Z" },
+    { url = "https://files.pythonhosted.org/packages/87/cb/ce35a65f83a47b510d8a2f1eddf3bdbb0d57aabc87351c8788caf3309f76/tree_sitter_python-0.23.6-cp39-abi3-win_arm64.whl", hash = "sha256:71334371bd73d5fe080aed39fbff49ed8efb9506edebe16795b0c7567ed6a272", size = 73649, upload-time = "2024-12-22T23:09:53.71Z" },
+]
+
+[[package]]
+name = "tree-sitter-typescript"
+version = "0.23.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/1e/fc/bb52958f7e399250aee093751e9373a6311cadbe76b6e0d109b853757f35/tree_sitter_typescript-0.23.2.tar.gz", hash = "sha256:7b167b5827c882261cb7a50dfa0fb567975f9b315e87ed87ad0a0a3aedb3834d", size = 773053, upload-time = "2024-11-11T02:36:11.396Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/28/95/4c00680866280e008e81dd621fd4d3f54aa3dad1b76b857a19da1b2cc426/tree_sitter_typescript-0.23.2-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:3cd752d70d8e5371fdac6a9a4df9d8924b63b6998d268586f7d374c9fba2a478", size = 286677, upload-time = "2024-11-11T02:35:58.839Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/2f/1f36fda564518d84593f2740d5905ac127d590baf5c5753cef2a88a89c15/tree_sitter_typescript-0.23.2-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:c7cc1b0ff5d91bac863b0e38b1578d5505e718156c9db577c8baea2557f66de8", size = 302008, upload-time = "2024-11-11T02:36:00.733Z" },
+    { url = "https://files.pythonhosted.org/packages/96/2d/975c2dad292aa9994f982eb0b69cc6fda0223e4b6c4ea714550477d8ec3a/tree_sitter_typescript-0.23.2-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4b1eed5b0b3a8134e86126b00b743d667ec27c63fc9de1b7bb23168803879e31", size = 351987, upload-time = "2024-11-11T02:36:02.669Z" },
+    { url = "https://files.pythonhosted.org/packages/49/d1/a71c36da6e2b8a4ed5e2970819b86ef13ba77ac40d9e333cb17df6a2c5db/tree_sitter_typescript-0.23.2-cp39-abi3-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e96d36b85bcacdeb8ff5c2618d75593ef12ebaf1b4eace3477e2bdb2abb1752c", size = 344960, upload-time = "2024-11-11T02:36:04.443Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/cb/f57b149d7beed1a85b8266d0c60ebe4c46e79c9ba56bc17b898e17daf88e/tree_sitter_typescript-0.23.2-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:8d4f0f9bcb61ad7b7509d49a1565ff2cc363863644a234e1e0fe10960e55aea0", size = 340245, upload-time = "2024-11-11T02:36:06.473Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/ab/dd84f0e2337296a5f09749f7b5483215d75c8fa9e33738522e5ed81f7254/tree_sitter_typescript-0.23.2-cp39-abi3-win_amd64.whl", hash = "sha256:3f730b66396bc3e11811e4465c41ee45d9e9edd6de355a58bbbc49fa770da8f9", size = 278015, upload-time = "2024-11-11T02:36:07.631Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/e4/81f9a935789233cf412a0ed5fe04c883841d2c8fb0b7e075958a35c65032/tree_sitter_typescript-0.23.2-cp39-abi3-win_arm64.whl", hash = "sha256:05db58f70b95ef0ea126db5560f3775692f609589ed6f8dd0af84b7f19f1cbb7", size = 274052, upload-time = "2024-11-11T02:36:09.514Z" },
+]
+
 [[package]]
 name = "triton"
 version = "3.3.1"