diff --git a/CHANGELOG.md b/CHANGELOG.md
index 4a0f8eb2..76f12654 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -16,6 +16,7 @@
   - Research graph receives compact context (~1,000-2,000 tokens) instead of raw qa_history (potentially thousands of tokens)
   - New `session_context` field on `ChatSessionState` synced via AG-UI state protocol
   - Chat TUI: New context modal (`Ctrl+O`) to view current session context
+  - Planner can now return empty sub_questions when context is sufficient to answer directly
 
 ### Changed
 
@@ -28,6 +29,13 @@
   - Removed `haiku.rag.embeddings.voyageai` module
   - The `voyageai` extra now delegates to `pydantic-ai-slim[voyageai]`
 
+### Removed
+
+- **Q&A History Functions**: Removed unused conversation history utilities
+  - `rank_qa_history_by_similarity()` - chat agent now uses `SessionContext` instead of ranked Q&A pairs
+  - `format_conversation_context()` - no longer needed with SessionContext approach
+  - Associated embedding cache and helper functions also removed
+
 ## [0.26.9] - 2026-01-22
 
 ### Fixed
@@ -35,6 +43,9 @@
 - **v0.25.0 Migration Failure**: Fixed "Table 'documents' already exists" error during migration caused by held table references preventing `drop_table()` from succeeding. Added recovery logic to restore documents from staging table if a previous migration attempt failed mid-way.
 
 ## [0.26.8] - 2026-01-22
+
+### Added
+
 - **Jina Reranker v3**: Added support for Jina reranking with API mode (`provider: jina`) and local inference (`provider: jina-local`, requires `[jina]` extra)
 - **Model Downloads**: `download-models` now pre-downloads HuggingFace models for `sentence-transformers`, `mxbai`, and `jina-local`
 - **Reranker Factory**: Removed unreliable `id(config)`-based caching from `get_reranker()`; factory now always instantiates fresh
diff --git a/haiku_rag_slim/haiku/rag/agents/chat/__init__.py b/haiku_rag_slim/haiku/rag/agents/chat/__init__.py
index dc55dc58..32356e76 100644
--- a/haiku_rag_slim/haiku/rag/agents/chat/__init__.py
+++ b/haiku_rag_slim/haiku/rag/agents/chat/__init__.py
@@ -10,7 +10,6 @@ from haiku.rag.agents.chat.state import (
     SearchDeps,
     SessionContext,
     build_document_filter,
-    format_conversation_context,
 )
 
 __all__ = [
@@ -24,7 +23,6 @@ __all__ = [
     "SearchDeps",
     "SessionContext",
     "build_document_filter",
-    "format_conversation_context",
     "summarize_session",
     "update_session_context",
 ]
diff --git a/haiku_rag_slim/haiku/rag/agents/chat/agent.py b/haiku_rag_slim/haiku/rag/agents/chat/agent.py
index 1e1fcc01..708e61ee 100644
--- a/haiku_rag_slim/haiku/rag/agents/chat/agent.py
+++ b/haiku_rag_slim/haiku/rag/agents/chat/agent.py
@@ -14,11 +14,9 @@ from haiku.rag.agents.chat.state import (
     CitationInfo,
     QAResponse,
     build_document_filter,
-    rank_qa_history_by_similarity,
 )
 from haiku.rag.agents.research.dependencies import ResearchContext
 from haiku.rag.agents.research.graph import build_conversational_graph
-from haiku.rag.agents.research.models import Citation, SearchAnswer
 from haiku.rag.agents.research.state import ResearchDeps, ResearchState
 from haiku.rag.config.models import AppConfig
 from haiku.rag.utils import get_model
@@ -181,61 +179,19 @@ def create_chat_agent(config: AppConfig) -> Agent[ChatDeps, str]:
         # Build filter from document_name
         doc_filter = build_document_filter(document_name) if document_name else None
 
-        # Filter and rank qa_history
-        ranked_history: list[QAResponse] = []
-        if ctx.deps.session_state and ctx.deps.session_state.qa_history:
-            # Step 1: Filter out low-confidence responses
-            filtered_history = [
-                qa for qa in ctx.deps.session_state.qa_history if qa.confidence >= 0.3
-            ]
-
-            # Step 2: Rank filtered history by similarity to current question
-            embedder = ctx.deps.client.chunk_repository.embedder
-            ranked_history = await rank_qa_history_by_similarity(
-                current_question=question,
-                qa_history=filtered_history,
-                embedder=embedder,
-                top_k=5,
-            )
-
-        # Convert ranked qa_history to SearchAnswers for context seeding
-        existing_qa: list[SearchAnswer] = []
-        if ranked_history:
-            for qa in ranked_history:
-                citations = [
-                    Citation(
-                        document_id=c.document_id,
-                        chunk_id=c.chunk_id,
-                        document_uri=c.document_uri,
-                        document_title=c.document_title,
-                        page_numbers=c.page_numbers,
-                        headings=c.headings,
-                        content=c.content,
-                    )
-                    for c in qa.citations
-                ]
-                existing_qa.append(
-                    SearchAnswer(
-                        query=qa.question,
-                        answer=qa.answer,
-                        confidence=qa.confidence,
-                        cited_chunks=[c.chunk_id for c in qa.citations],
-                        citations=citations,
-                    )
-                )
-
         # Build and run the conversational research graph
         graph = build_conversational_graph(config=ctx.deps.config)
 
-        # Determine background context:
-        # 1. Use session_context summary if available (compressed history)
-        # 2. Fall back to explicit background_context if set
+        # Determine context strategy:
+        # 1. If session_context exists, use compressed summary (skip raw qa_history)
+        # 2. Otherwise, fall back to explicit background_context
         background_context: str | None = None
         if ctx.deps.session_state:
             if (
                 ctx.deps.session_state.session_context
                 and ctx.deps.session_state.session_context.summary
             ):
+                # Use compressed SessionContext - no need for raw qa_history
                 background_context = (
                     ctx.deps.session_state.session_context.render_markdown()
                 )
@@ -244,7 +200,6 @@ def create_chat_agent(config: AppConfig) -> Agent[ChatDeps, str]:
 
         context = ResearchContext(
             original_question=question,
-            qa_responses=existing_qa,
             background_context=background_context,
         )
         state = ResearchState(
diff --git a/haiku_rag_slim/haiku/rag/agents/chat/state.py b/haiku_rag_slim/haiku/rag/agents/chat/state.py
index 0f0d4c86..40af1fd9 100644
--- a/haiku_rag_slim/haiku/rag/agents/chat/state.py
+++ b/haiku_rag_slim/haiku/rag/agents/chat/state.py
@@ -1,31 +1,17 @@
-import hashlib
 from dataclasses import dataclass, field
 from datetime import datetime
-from typing import TYPE_CHECKING, Any
+from typing import Any
 
-import numpy as np
-from numpy.typing import NDArray
 from pydantic import BaseModel
-from pydantic_ai import format_as_xml
 
 from haiku.rag.client import HaikuRAG
 from haiku.rag.config.models import AppConfig
 from haiku.rag.store.models import SearchResult
 
-if TYPE_CHECKING:
-    from haiku.rag.embeddings import EmbedderWrapper
-
 MAX_QA_HISTORY = 50
 
 AGUI_STATE_KEY = "haiku.rag.chat"
 
-_embedding_cache: dict[str, list[float]] = {}
-
-
-def _qa_cache_key(question: str, answer: str) -> str:
-    """Generate cache key from Q/A content."""
-    return hashlib.sha256(f"Q: {question}\nA: {answer}".encode()).hexdigest()
-
 
 class CitationInfo(BaseModel):
     """Citation info for frontend display."""
@@ -64,9 +50,7 @@ class SessionContext(BaseModel):
 
     def render_markdown(self) -> str:
         """Render context for injection into research graph."""
-        if not self.summary:
-            return ""
-        return f"## Prior Conversation Context\n\n{self.summary}"
+        return self.summary
 
 
 class ChatSessionState(BaseModel):
@@ -79,96 +63,6 @@ class ChatSessionState(BaseModel):
     session_context: SessionContext | None = None
 
 
-def format_conversation_context(qa_history: list[QAResponse]) -> str:
-    """Format conversation history as XML for inclusion in prompts."""
-    if not qa_history:
-        return ""
-
-    context_data = {
-        "previous_qa": [
-            {
-                "question": qa.question,
-                "answer": qa.answer,
-                "sources": qa.sources,
-            }
-            for qa in qa_history
-        ],
-    }
-    return format_as_xml(context_data, root_tag="conversation_context")
-
-
-def _cosine_similarity(a: NDArray[np.float64], b: NDArray[np.float64]) -> float:
-    """Compute cosine similarity between two vectors."""
-    return float(np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b)))
-
-
-async def rank_qa_history_by_similarity(
-    current_question: str,
-    qa_history: list[QAResponse],
-    embedder: "EmbedderWrapper",
-    top_k: int = 5,
-) -> list[QAResponse]:
-    """Rank Q&A history by semantic similarity to current question.
-
-    Embeds question+answer pairs and returns the top-K most similar to the
-    current question. Falls back to returning the last top_k entries if
-    embedding fails.
-
-    Args:
-        current_question: The current question to compare against.
-        qa_history: List of previous Q&A pairs.
-        embedder: Embedder instance to use for embedding.
-        top_k: Maximum number of entries to return.
-
-    Returns:
-        Top-K Q&A pairs ranked by similarity to current question.
-    """
-    if not qa_history:
-        return []
-
-    if len(qa_history) <= top_k:
-        return qa_history
-
-    # Embed current question
-    question_embedding = np.array(await embedder.embed_query(current_question))
-
-    # Check cache and collect uncached entries
-    qa_embeddings: list[list[float]] = []
-    uncached_indices: list[int] = []
-    uncached_texts: list[str] = []
-
-    for i, qa in enumerate(qa_history):
-        cache_key = _qa_cache_key(qa.question, qa.answer)
-        if cache_key in _embedding_cache:
-            qa_embeddings.append(_embedding_cache[cache_key])
-        else:
-            qa_embeddings.append([])  # placeholder
-            uncached_indices.append(i)
-            uncached_texts.append(f"Q: {qa.question}\nA: {qa.answer}")
-
-    # Embed only uncached entries
-    if uncached_texts:
-        new_embeddings = await embedder.embed_documents(uncached_texts)
-        for idx, embedding in zip(uncached_indices, new_embeddings):
-            qa = qa_history[idx]
-            cache_key = _qa_cache_key(qa.question, qa.answer)
-            _embedding_cache[cache_key] = embedding
-            qa_embeddings[idx] = embedding
-
-    # Compute similarities
-    similarities: list[tuple[int, float]] = []
-    for i, qa_emb in enumerate(qa_embeddings):
-        sim = _cosine_similarity(question_embedding, np.array(qa_emb))
-        similarities.append((i, sim))
-
-    # Sort by similarity (descending) and take top-K
-    similarities.sort(key=lambda x: x[1], reverse=True)
-    top_indices = sorted([idx for idx, _ in similarities[:top_k]])
-
-    # Return in original order
-    return [qa_history[i] for i in top_indices]
-
-
 @dataclass
 class ChatDeps:
     """Dependencies for chat agent.
diff --git a/haiku_rag_slim/haiku/rag/agents/research/models.py b/haiku_rag_slim/haiku/rag/agents/research/models.py
index 9cd37168..b58a4f81 100644
--- a/haiku_rag_slim/haiku/rag/agents/research/models.py
+++ b/haiku_rag_slim/haiku/rag/agents/research/models.py
@@ -17,8 +17,6 @@ class ResearchPlan(BaseModel):
     @field_validator("sub_questions")
     @classmethod
     def validate_sub_questions(cls, v: list[str]) -> list[str]:
-        if len(v) < 1:
-            raise ValueError("Must have at least 1 sub-question")
         if len(v) > 12:
             raise ValueError("Cannot have more than 12 sub-questions")
         return v
diff --git a/tests/agents/chat/test_chat_agent.py b/tests/agents/chat/test_chat_agent.py
index 9eb05cce..e2c29cf0 100644
--- a/tests/agents/chat/test_chat_agent.py
+++ b/tests/agents/chat/test_chat_agent.py
@@ -201,75 +201,6 @@ Scanned documents were excluded to avoid rotation and skewing issues.
 """
 
 
-@pytest.mark.asyncio
-@pytest.mark.vcr()
-async def test_chat_agent_with_qa_history_ranking(allow_model_requests, temp_db_path):
-    """Test chat agent uses similarity ranking for qa_history.
-
-    This test verifies that when qa_history has more than 5 entries,
-    the ranking function is applied and the agent can still process requests.
-    """
-    async with HaikuRAG(temp_db_path, create=True) as client:
-        # Add a simple document
-        await client.create_document(
-            content=DOCLAYNET_CLASS_LABELS,
-            uri="doclaynet-labels",
-            title="DocLayNet Class Labels",
-        )
-
-        agent = create_chat_agent(Config)
-
-        # Build session state with pre-populated qa_history (>5 items to trigger ranking)
-        session_state = ChatSessionState(
-            session_id="test-ranking",
-            qa_history=[
-                QAResponse(
-                    question="What are the 11 class labels in DocLayNet?",
-                    answer="The 11 class labels are: Caption, Footnote, Formula, List-item, Page-footer, Page-header, Picture, Section-header, Table, Text, and Title.",
-                ),
-                QAResponse(
-                    question="How was the annotation process organized?",
-                    answer="The annotation was organized into 4 phases.",
-                ),
-                QAResponse(
-                    question="What data sources were used?",
-                    answer="Sources include arXiv and government offices.",
-                ),
-                QAResponse(
-                    question="How were pages selected?",
-                    answer="By selective subsampling.",
-                ),
-                QAResponse(
-                    question="What is the agreement metric?",
-                    answer="The mAP metric was used.",
-                ),
-                QAResponse(
-                    question="What is machine learning?",
-                    answer="A field of AI.",
-                ),
-            ],
-        )
-
-        deps = ChatDeps(
-            client=client,
-            config=Config,
-            session_state=session_state,
-        )
-
-        # Ask a question - the key test is that ranking is applied without error
-        result = await agent.run(
-            "What class labels are defined in the dataset?",
-            deps=deps,
-        )
-
-        # Verify the agent produced a response (ranking didn't break anything)
-        assert result.output is not None
-        assert len(result.output) > 0
-
-        # Verify qa_history was updated (new Q&A was added)
-        assert len(session_state.qa_history) == 7  # 6 original + 1 new
-
-
 @pytest.mark.asyncio
 @pytest.mark.vcr()
 async def test_chat_agent_search_tool(allow_model_requests, temp_db_path):
diff --git a/tests/agents/chat/test_context.py b/tests/agents/chat/test_context.py
index 90ac3e9e..9906e8e9 100644
--- a/tests/agents/chat/test_context.py
+++ b/tests/agents/chat/test_context.py
@@ -41,14 +41,10 @@ class TestSessionContext:
         assert ctx.render_markdown() == ""
 
     def test_render_markdown_with_summary(self):
-        """Test render_markdown returns formatted markdown."""
-        ctx = SessionContext(
-            summary="## Key Facts\n- Authentication uses JWT\n- Rate limit is 100/min"
-        )
-        result = ctx.render_markdown()
-        assert "## Prior Conversation Context" in result
-        assert "Authentication uses JWT" in result
-        assert "Rate limit is 100/min" in result
+        """Test render_markdown returns the summary directly."""
+        summary = "## Key Facts\n- Authentication uses JWT\n- Rate limit is 100/min"
+        ctx = SessionContext(summary=summary)
+        assert ctx.render_markdown() == summary
 
     def test_session_context_serialization_roundtrip(self):
         """Test SessionContext serializes and deserializes correctly."""
diff --git a/tests/agents/chat/test_state.py b/tests/agents/chat/test_state.py
index b5fef40f..fd2a4df2 100644
--- a/tests/agents/chat/test_state.py
+++ b/tests/agents/chat/test_state.py
@@ -1,224 +1,8 @@
-from pathlib import Path
-
-import pytest
-
 from haiku.rag.agents.chat.state import (
     MAX_QA_HISTORY,
-    CitationInfo,
     QAResponse,
-    _embedding_cache,
-    _qa_cache_key,
     build_document_filter,
-    format_conversation_context,
-    rank_qa_history_by_similarity,
 )
-from haiku.rag.client import HaikuRAG
-
-
-@pytest.fixture(scope="module")
-def vcr_cassette_dir():
-    return str(Path(__file__).parent.parent.parent / "cassettes" / "test_chat_state")
-
-
-@pytest.mark.asyncio
-async def test_rank_qa_history_empty():
-    """Test empty history returns empty list."""
-    # Create a mock embedder - we won't actually call it
-    result = await rank_qa_history_by_similarity(
-        current_question="What is this?",
-        qa_history=[],
-        embedder=None,  # type: ignore - won't be called for empty list
-        top_k=5,
-    )
-    assert result == []
-
-
-@pytest.mark.asyncio
-@pytest.mark.vcr()
-async def test_rank_qa_history_small_list(temp_db_path, allow_model_requests):
-    """Test with history smaller than top_k returns all entries."""
-    async with HaikuRAG(temp_db_path, create=True) as client:
-        embedder = client.chunk_repository.embedder
-
-        # Create 3 Q&A pairs (less than top_k=5)
-        qa_history = [
-            QAResponse(question="What is Python?", answer="A programming language"),
-            QAResponse(question="What is Java?", answer="Another programming language"),
-            QAResponse(
-                question="What is Rust?", answer="A systems programming language"
-            ),
-        ]
-
-        result = await rank_qa_history_by_similarity(
-            current_question="Tell me about Python",
-            qa_history=qa_history,
-            embedder=embedder,
-            top_k=5,
-        )
-
-        # Should return all 3 entries since history < top_k
-        assert len(result) == 3
-        # All original entries should be present
-        assert set(qa.question for qa in result) == {
-            "What is Python?",
-            "What is Java?",
-            "What is Rust?",
-        }
-
-
-@pytest.mark.asyncio
-@pytest.mark.vcr()
-async def test_rank_qa_history_returns_top_k(temp_db_path, allow_model_requests):
-    """Test ranking returns top-K most similar entries and populates cache."""
-    async with HaikuRAG(temp_db_path, create=True) as client:
-        embedder = client.chunk_repository.embedder
-
-        # Create 10 Q&A pairs on different topics
-        qa_history = [
-            QAResponse(
-                question="What are the 11 class labels in DocLayNet?",
-                answer="Caption, Footnote, Formula, List-item, Page-footer, Page-header, Picture, Section-header, Table, Text, and Title",
-            ),
-            QAResponse(
-                question="How was the annotation process organized?",
-                answer="The process had 4 phases with 40 dedicated annotators",
-            ),
-            QAResponse(
-                question="What data sources were used?",
-                answer="arXiv, government offices, company websites, financial reports and patents",
-            ),
-            QAResponse(
-                question="How were pages selected?",
-                answer="By selective subsampling with bias towards pages with figures or tables",
-            ),
-            QAResponse(
-                question="What is the inter-annotator agreement?",
-                answer="Computed as mAP@0.5-0.95 metric between pairwise annotations",
-            ),
-            QAResponse(
-                question="What is machine learning?",
-                answer="A field of AI that enables systems to learn from data",
-            ),
-            QAResponse(
-                question="How does neural network training work?",
-                answer="Through backpropagation and gradient descent",
-            ),
-            QAResponse(
-                question="What is deep learning?",
-                answer="A subset of ML using neural networks with many layers",
-            ),
-        ]
-
-        # Clear cache to verify it gets populated
-        _embedding_cache.clear()
-
-        # Ask a question related to class labels (Q1)
-        result = await rank_qa_history_by_similarity(
-            current_question="Which class label has the highest count in DocLayNet?",
-            qa_history=qa_history,
-            embedder=embedder,
-            top_k=5,
-        )
-
-        # Should return exactly 5 entries
-        assert len(result) == 5
-
-        # The class labels Q&A should be in the top 5 (it's most semantically similar)
-        result_questions = [qa.question for qa in result]
-        assert "What are the 11 class labels in DocLayNet?" in result_questions
-
-        # Verify cache is populated for all Q/A pairs
-        for qa in qa_history:
-            cache_key = _qa_cache_key(qa.question, qa.answer)
-            assert cache_key in _embedding_cache
-            assert len(_embedding_cache[cache_key]) > 0
-
-
-@pytest.mark.asyncio
-@pytest.mark.vcr()
-async def test_rank_qa_history_preserves_order(temp_db_path, allow_model_requests):
-    """Test that ranking preserves original order among selected items."""
-    async with HaikuRAG(temp_db_path, create=True) as client:
-        embedder = client.chunk_repository.embedder
-
-        # Create Q&A pairs where multiple are similar
-        qa_history = [
-            QAResponse(
-                question="What is Python?",
-                answer="A programming language",
-                citations=[
-                    CitationInfo(
-                        index=1,
-                        document_id="doc1",
-                        chunk_id="chunk1",
-                        document_uri="python.md",
-                        content="Python content",
-                    )
-                ],
-            ),
-            QAResponse(
-                question="What is Java?",
-                answer="Another programming language",
-            ),
-            QAResponse(
-                question="How to use Python for data science?",
-                answer="Use pandas, numpy, and scikit-learn",
-            ),
-        ]
-
-        result = await rank_qa_history_by_similarity(
-            current_question="Tell me about Python programming",
-            qa_history=qa_history,
-            embedder=embedder,
-            top_k=3,
-        )
-
-        # All should be returned
-        assert len(result) == 3
-
-        # The two Python-related questions should be in the results
-        result_questions = [qa.question for qa in result]
-        assert "What is Python?" in result_questions
-        assert "How to use Python for data science?" in result_questions
-
-
-def test_format_conversation_context_empty():
-    """Test format_conversation_context with empty history."""
-    result = format_conversation_context([])
-    assert result == ""
-
-
-def test_format_conversation_context_with_history():
-    """Test format_conversation_context formats qa_history as XML."""
-    citation = CitationInfo(
-        index=1,
-        document_id="doc-123",
-        chunk_id="chunk-456",
-        document_uri="test.md",
-        document_title="Test Document",
-        content="Test content",
-    )
-    qa_history = [
-        QAResponse(
-            question="What is Python?",
-            answer="A programming language",
-            citations=[citation],
-        ),
-        QAResponse(
-            question="What is Java?",
-            answer="Another programming language",
-        ),
-    ]
-
-    result = format_conversation_context(qa_history)
-
-    assert "<conversation_context>" in result
-    assert "previous_qa" in result
-    assert "What is Python?" in result
-    assert "A programming language" in result
-    assert "What is Java?" in result
-    assert "Another programming language" in result
-    assert "Test Document" in result  # source from first citation
 
 
 def test_build_document_filter_simple():
diff --git a/tests/agents/research/test_research_graph.py b/tests/agents/research/test_research_graph.py
index b080d8f8..efe1d1e7 100644
--- a/tests/agents/research/test_research_graph.py
+++ b/tests/agents/research/test_research_graph.py
@@ -103,3 +103,21 @@ def test_format_conversational_context_for_prompt_excludes_background_when_none(
     context = ResearchContext(original_question="What is X?")
     result = format_conversational_context_for_prompt(context)
     assert "<background>" not in result
+
+
+def test_research_plan_allows_empty_sub_questions():
+    """Test ResearchPlan accepts empty sub_questions when context is sufficient."""
+    from haiku.rag.agents.research.models import ResearchPlan
+
+    plan = ResearchPlan(sub_questions=[])
+    assert plan.sub_questions == []
+
+
+def test_research_plan_rejects_too_many_sub_questions():
+    """Test ResearchPlan rejects more than 12 sub_questions."""
+    from pydantic import ValidationError
+
+    from haiku.rag.agents.research.models import ResearchPlan
+
+    with pytest.raises(ValidationError, match="Cannot have more than 12"):
+        ResearchPlan(sub_questions=[f"q{i}" for i in range(13)])
diff --git a/tests/cassettes/test_chat_agent/test_chat_agent_with_qa_history_ranking.yaml b/tests/cassettes/test_chat_agent/test_chat_agent_with_qa_history_ranking.yaml
deleted file mode 100644
index 1f2dd239..00000000
--- a/tests/cassettes/test_chat_agent/test_chat_agent_with_qa_history_ranking.yaml
+++ /dev/null
@@ -1,1581 +0,0 @@
-interactions:
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '730'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      encoding_format: base64
-      input:
-      - |-
-        DocLayNet Dataset - Class Labels
-        DocLayNet defines 11 distinct class labels for document layout analysis:
-        1. Caption - Text describing figures or tables
-        2. Footnote - Notes at the bottom of pages
-        3. Formula - Mathematical expressions
-        4. List-item - Items in bulleted or numbered lists
-        5. Page-footer - Footer content on pages
-        6. Page-header - Header content on pages
-        7. Picture - Images and diagrams
-        8. Section-header - Headings for document sections
-        9. Table - Tabular data
-        10. Text - Regular paragraph text (highest count: 510,377 instances)
-        11. Title - Document titles
-        The Text class has the highest count with 510,377 instances in the dataset.
-      model: qwen3-embedding:4b
-    uri: http://localhost:11434/v1/embeddings
-  response:
-    headers:
-      content-type:
-      - application/json
-      transfer-encoding:
-      - chunked
-    parsed_body:
-      data:
-      - embedding: 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
-        index: 0
-        object: embedding
-      model: qwen3-embedding:4b
-      object: list
-      usage:
-        prompt_tokens: 166
-        total_tokens: 166
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '4087'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          You are a helpful research assistant powered by haiku.rag, a knowledge base system.
-
-          You have access to a knowledge base of documents. Use your tools to search and answer questions.
-
-          CRITICAL RULES:
-          1. For greetings or casual chat: respond directly WITHOUT using any tools
-          2. For questions: Use the "ask" tool EXACTLY ONCE - it handles query expansion internally
-          3. For searches: Use the "search" tool EXACTLY ONCE - it handles multi-query expansion internally
-          4. NEVER call the same tool multiple times for a single user message
-          5. NEVER make up information - always use tools to get facts from the knowledge base
-
-          How to decide which tool to use:
-          - "get_document" - Use when the user references a SPECIFIC document by name, title, or URI (e.g., "summarize document X", "get the paper about Y", "fetch 2412.00566"). Retrieves the full document content.
-          - "ask" - Use for general questions about topics in the knowledge base when no specific document is named. It searches across all documents and returns answers with citations.
-          - "search" - Use when the user explicitly asks to search/find/explore documents. Call it ONCE. After calling search, copy the ENTIRE tool response to your output INCLUDING the content snippets. Do NOT shorten, summarize, or omit any part of the results.
-
-          IMPORTANT - When user mentions a document in search/ask:
-          - If user says "search in <doc>", "find in <doc>", "answer from <doc>", or "<topic> in <doc>":
-            - Extract the TOPIC as `query`/`question`
-            - Extract the DOCUMENT NAME as `document_name`
-          - Examples for search:
-            - "search for embeddings in the ML paper" → query="embeddings", document_name="ML paper"
-            - "find transformer architecture in 2412.00566" → query="transformer architecture", document_name="2412.00566"
-          - Examples for ask:
-            - "what does the ML paper say about embeddings?" → question="what are the embedding methods?", document_name="ML paper"
-            - "answer from 2412.00566 about model training" → question="how is the model trained?", document_name="2412.00566"
-
-          Be friendly and conversational. When you use the "ask" tool, summarize the key findings for the user.
-        role: system
-      - content: What class labels are defined in the dataset?
-        role: user
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: |-
-            Search the knowledge base for relevant documents.
-
-            Use this when you need to find documents or explore the knowledge base.
-            Results are displayed to the user - just list the titles found.
-          name: search
-          parameters:
-            additionalProperties: false
-            properties:
-              document_name:
-                anyOf:
-                - type: string
-                - type: 'null'
-                default: null
-                description: Optional document name/title to search within
-              limit:
-                anyOf:
-                - type: integer
-                - type: 'null'
-                default: null
-                description: 'Number of results to return (default: 5)'
-              query:
-                description: The search query (what to search for)
-                type: string
-            required:
-            - query
-            type: object
-        type: function
-      - function:
-          description: |-
-            Answer a specific question using the knowledge base.
-
-            Use this for direct questions that need a focused answer with citations.
-            Uses a research graph for planning, searching, and synthesis.
-          name: ask
-          parameters:
-            additionalProperties: false
-            properties:
-              document_name:
-                anyOf:
-                - type: string
-                - type: 'null'
-                default: null
-                description: Optional document name/title to search within (e.g., "tbmed593", "army manual")
-              question:
-                description: The question to answer
-                type: string
-            required:
-            - question
-            type: object
-        type: function
-      - function:
-          description: |-
-            Retrieve a specific document by title or URI.
-
-            Use this when the user wants to fetch/get/retrieve a specific document.
-          name: get_document
-          parameters:
-            additionalProperties: false
-            properties:
-              query:
-                description: The document title or URI to look up
-                type: string
-            required:
-            - query
-            type: object
-          strict: true
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '556'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: tool_calls
-        index: 0
-        message:
-          content: ''
-          reasoning: User asks about class labels defined in dataset. Use ask.
-          role: assistant
-          tool_calls:
-          - function:
-              arguments: '{"document_name":null,"question":"What class labels are defined in the dataset?"}'
-              name: ask
-            id: call_hcs4zx6t
-            index: 0
-            type: function
-      created: 1768998207
-      id: chatcmpl-101
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 48
-        prompt_tokens: 846
-        total_tokens: 894
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '115'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      encoding_format: base64
-      input:
-      - What class labels are defined in the dataset?
-      model: qwen3-embedding:4b
-    uri: http://localhost:11434/v1/embeddings
-  response:
-    headers:
-      content-type:
-      - application/json
-      transfer-encoding:
-      - chunked
-    parsed_body:
-      data:
-      - embedding: 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
-        index: 0
-        object: embedding
-      model: qwen3-embedding:4b
-      object: list
-      usage:
-        prompt_tokens: 10
-        total_tokens: 10
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '611'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      encoding_format: base64
-      input:
-      - |-
-        Q: What are the 11 class labels in DocLayNet?
-        A: The 11 class labels are: Caption, Footnote, Formula, List-item, Page-footer, Page-header, Picture, Section-header, Table, Text, and Title.
-      - |-
-        Q: How was the annotation process organized?
-        A: The annotation was organized into 4 phases.
-      - |-
-        Q: What data sources were used?
-        A: Sources include arXiv and government offices.
-      - |-
-        Q: How were pages selected?
-        A: By selective subsampling.
-      - |-
-        Q: What is the agreement metric?
-        A: The mAP metric was used.
-      - |-
-        Q: What is machine learning?
-        A: A field of AI.
-      model: qwen3-embedding:4b
-    uri: http://localhost:11434/v1/embeddings
-  response:
-    headers:
-      content-type:
-      - application/json
-      transfer-encoding:
-      - chunked
-    parsed_body:
-      data:
-      - embedding: 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
-        index: 0
-        object: embedding
-      - embedding: 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
-        index: 1
-        object: embedding
-      - embedding: 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
-        index: 2
-        object: embedding
-      - embedding: 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
-        index: 3
-        object: embedding
-      - embedding: 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
-        index: 4
-        object: embedding
-      - embedding: 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
-        index: 5
-        object: embedding
-      model: qwen3-embedding:4b
-      object: list
-      usage:
-        prompt_tokens: 144
-        total_tokens: 144
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '3223'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          You are the research orchestrator for a focused workflow.
-
-          You have access to context that may include:
-          - <background>: Domain context for the conversation
-          - <prior_answers>: Previous Q&A pairs with confidence scores
-
-          Review this first - if prior answers already answer the question completely,
-          you may return an empty sub_questions list. Only create sub-questions to
-          fill genuine gaps.
-
-          Responsibilities:
-          1. Review prior_answers to understand what's already known
-          2. Identify gaps that need additional research
-          3. Propose minimal sub-questions only for missing information
-
-          Plan requirements:
-          - If prior answers fully answer the question, return an empty sub_questions list.
-          - Only create new sub-questions for genuine gaps in existing knowledge.
-          - sub_questions must be a list of plain strings (max 3).
-          - Each sub_question must be standalone and self-contained.
-          - Prioritize the highest-value gaps first.
-
-          Use the gather_context tool once on the main question before planning.
-        role: system
-      - content: |-
-          Review existing context and plan additional research if needed.
-
-          <context>
-            <question>What class labels are defined in the dataset?</question>
-            <prior_answers>
-              <item>
-                <question>What are the 11 class labels in DocLayNet?</question>
-                <answer>The 11 class labels are: Caption, Footnote, Formula, List-item, Page-footer, Page-header, Picture, Section-header, Table, Text, and Title.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>How was the annotation process organized?</question>
-                <answer>The annotation was organized into 4 phases.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What data sources were used?</question>
-                <answer>Sources include arXiv and government offices.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What is the agreement metric?</question>
-                <answer>The mAP metric was used.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What is machine learning?</question>
-                <answer>A field of AI.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-            </prior_answers>
-          </context>
-
-          Main question: What class labels are defined in the dataset?
-        role: user
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: ''
-          name: gather_context
-          parameters:
-            additionalProperties: false
-            properties:
-              limit:
-                anyOf:
-                - type: integer
-                - type: 'null'
-                default: null
-              query:
-                type: string
-            required:
-            - query
-            type: object
-        type: function
-      - function:
-          description: A structured research plan with sub-questions to explore.
-          name: final_result
-          parameters:
-            additionalProperties: false
-            properties:
-              sub_questions:
-                description: Specific questions to research, phrased as complete questions
-                items:
-                  type: string
-                type: array
-            required:
-            - sub_questions
-            type: object
-          strict: true
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '405'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: stop
-        index: 0
-        message:
-          content: '{"sub_questions":[] }'
-          reasoning: We have prior answer listing 11 labels. That seems complete. So no further research.
-          role: assistant
-      created: 1768998209
-      id: chatcmpl-51
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 37
-        prompt_tokens: 692
-        total_tokens: 729
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '3664'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          You are the research orchestrator for a focused workflow.
-
-          You have access to context that may include:
-          - <background>: Domain context for the conversation
-          - <prior_answers>: Previous Q&A pairs with confidence scores
-
-          Review this first - if prior answers already answer the question completely,
-          you may return an empty sub_questions list. Only create sub-questions to
-          fill genuine gaps.
-
-          Responsibilities:
-          1. Review prior_answers to understand what's already known
-          2. Identify gaps that need additional research
-          3. Propose minimal sub-questions only for missing information
-
-          Plan requirements:
-          - If prior answers fully answer the question, return an empty sub_questions list.
-          - Only create new sub-questions for genuine gaps in existing knowledge.
-          - sub_questions must be a list of plain strings (max 3).
-          - Each sub_question must be standalone and self-contained.
-          - Prioritize the highest-value gaps first.
-
-          Use the gather_context tool once on the main question before planning.
-        role: system
-      - content: |-
-          Review existing context and plan additional research if needed.
-
-          <context>
-            <question>What class labels are defined in the dataset?</question>
-            <prior_answers>
-              <item>
-                <question>What are the 11 class labels in DocLayNet?</question>
-                <answer>The 11 class labels are: Caption, Footnote, Formula, List-item, Page-footer, Page-header, Picture, Section-header, Table, Text, and Title.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>How was the annotation process organized?</question>
-                <answer>The annotation was organized into 4 phases.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What data sources were used?</question>
-                <answer>Sources include arXiv and government offices.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What is the agreement metric?</question>
-                <answer>The mAP metric was used.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What is machine learning?</question>
-                <answer>A field of AI.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-            </prior_answers>
-          </context>
-
-          Main question: What class labels are defined in the dataset?
-        role: user
-      - content: |-
-          <think>
-          We have prior answer listing 11 labels. That seems complete. So no further research.
-          </think>
-
-          {"sub_questions":[] }
-        role: assistant
-      - content: |-
-          1 validation error:
-          ```json
-          [
-            {
-              "type": "value_error",
-              "loc": [
-                "sub_questions"
-              ],
-              "msg": "Value error, Must have at least 1 sub-question",
-              "input": []
-            }
-          ]
-          ```
-
-          Fix the errors and try again.
-        role: user
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: ''
-          name: gather_context
-          parameters:
-            additionalProperties: false
-            properties:
-              limit:
-                anyOf:
-                - type: integer
-                - type: 'null'
-                default: null
-              query:
-                type: string
-            required:
-            - query
-            type: object
-        type: function
-      - function:
-          description: A structured research plan with sub-questions to explore.
-          name: final_result
-          parameters:
-            additionalProperties: false
-            properties:
-              sub_questions:
-                description: Specific questions to research, phrased as complete questions
-                items:
-                  type: string
-                type: array
-            required:
-            - sub_questions
-            type: object
-          strict: true
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '966'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: tool_calls
-        index: 0
-        message:
-          content: ''
-          reasoning: 'We need at least one sub-question, but we determined no gap. But tool requires at least 1. We can provide
-            a dummy question? But must be genuine gap. Maybe research context of other class labels? But we have. Could ask
-            about distribution? But main question already answered. We can still ask about any additional class labels beyond
-            11. Could query if any sub-classes? Provide sub-question: "Are there any class labels beyond the 11 listed for
-            DocLayNet?"'
-          role: assistant
-          tool_calls:
-          - function:
-              arguments: '{"sub_questions":["Are there any class labels beyond the 11 listed for DocLayNet?"]}'
-              name: final_result
-            id: call_bgwl6sb3
-            index: 0
-            type: function
-      created: 1768998211
-      id: chatcmpl-211
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 139
-        prompt_tokens: 795
-        total_tokens: 934
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '2852'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          You are a search and question-answering specialist.
-
-          Process:
-          1. Call search_and_answer with relevant keywords from the question.
-          2. Review the results ordered by relevance.
-          3. If needed, perform follow-up searches with different keywords (max 3 total).
-          4. Provide a concise answer based strictly on the retrieved content.
-
-          The search tool returns results like:
-          [9bde5847-44c9-400a-8997-0e6b65babf92] [rank 1 of 5]
-          Source: "Document Title" > Section > Subsection
-          Type: paragraph
-          Content:
-          The actual text content here...
-
-          [d5a63c82-cb40-439f-9b2e-de7d177829b7] [rank 2 of 5]
-          Source: "Another Document"
-          Type: table
-          Content:
-          | Column 1 | Column 2 |
-          ...
-
-          Each result includes:
-          - chunk_id in brackets and rank position (rank 1 = most relevant)
-          - Source: document title and section hierarchy (when available)
-          - Type: content type like paragraph, table, code, list_item (when available)
-          - Content: the actual text
-
-          Output format:
-          - query: Echo the question you are answering
-          - answer: Your concise answer based on the retrieved content
-          - cited_chunks: List of plain strings containing only the chunk UUIDs (not objects)
-          - confidence: A score from 0.0 to 1.0 indicating answer confidence
-
-          IMPORTANT: Use the EXACT, COMPLETE chunk ID (full UUID). Do NOT truncate IDs.
-
-          Guidelines:
-          - Base answers strictly on retrieved content - do not use external knowledge.
-          - Use the Source and Type metadata to understand context.
-          - If multiple results are relevant, synthesize them coherently.
-          - If information is insufficient, say so clearly.
-          - Be concise and direct; avoid meta commentary about the process.
-          - Results are ordered by relevance, with rank 1 being most relevant.
-        role: system
-      - content: Are there any class labels beyond the 11 listed for DocLayNet?
-        role: user
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: Search the knowledge base for relevant documents.
-          name: search_and_answer
-          parameters:
-            additionalProperties: false
-            properties:
-              limit:
-                anyOf:
-                - type: integer
-                - type: 'null'
-                default: null
-              query:
-                type: string
-            required:
-            - query
-            type: object
-        type: function
-      - function:
-          description: Answer to a search query with chunk references.
-          name: final_result
-          parameters:
-            additionalProperties: false
-            properties:
-              answer:
-                description: The answer to the question
-                type: string
-              cited_chunks:
-                description: IDs of chunks used to form the answer
-                items:
-                  type: string
-                type: array
-              confidence:
-                default: 1.0
-                description: Confidence score for this answer (0-1)
-                maximum: 1.0
-                minimum: 0.0
-                type: number
-              query:
-                description: The question that was answered
-                type: string
-            required:
-            - query
-            - answer
-            type: object
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '512'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: tool_calls
-        index: 0
-        message:
-          content: ''
-          reasoning: Need to search.
-          role: assistant
-          tool_calls:
-          - function:
-              arguments: '{"limit":5,"query":"DocLayNet class labels beyond the 11 listed"}'
-              name: search_and_answer
-            id: call_2dyiicy1
-            index: 0
-            type: function
-      created: 1768998213
-      id: chatcmpl-730
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 42
-        prompt_tokens: 630
-        total_tokens: 672
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '113'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      encoding_format: base64
-      input:
-      - DocLayNet class labels beyond the 11 listed
-      model: qwen3-embedding:4b
-    uri: http://localhost:11434/v1/embeddings
-  response:
-    headers:
-      content-type:
-      - application/json
-      transfer-encoding:
-      - chunked
-    parsed_body:
-      data:
-      - embedding: 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
-        index: 0
-        object: embedding
-      model: qwen3-embedding:4b
-      object: list
-      usage:
-        prompt_tokens: 13
-        total_tokens: 13
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '3720'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          You are a search and question-answering specialist.
-
-          Process:
-          1. Call search_and_answer with relevant keywords from the question.
-          2. Review the results ordered by relevance.
-          3. If needed, perform follow-up searches with different keywords (max 3 total).
-          4. Provide a concise answer based strictly on the retrieved content.
-
-          The search tool returns results like:
-          [9bde5847-44c9-400a-8997-0e6b65babf92] [rank 1 of 5]
-          Source: "Document Title" > Section > Subsection
-          Type: paragraph
-          Content:
-          The actual text content here...
-
-          [d5a63c82-cb40-439f-9b2e-de7d177829b7] [rank 2 of 5]
-          Source: "Another Document"
-          Type: table
-          Content:
-          | Column 1 | Column 2 |
-          ...
-
-          Each result includes:
-          - chunk_id in brackets and rank position (rank 1 = most relevant)
-          - Source: document title and section hierarchy (when available)
-          - Type: content type like paragraph, table, code, list_item (when available)
-          - Content: the actual text
-
-          Output format:
-          - query: Echo the question you are answering
-          - answer: Your concise answer based on the retrieved content
-          - cited_chunks: List of plain strings containing only the chunk UUIDs (not objects)
-          - confidence: A score from 0.0 to 1.0 indicating answer confidence
-
-          IMPORTANT: Use the EXACT, COMPLETE chunk ID (full UUID). Do NOT truncate IDs.
-
-          Guidelines:
-          - Base answers strictly on retrieved content - do not use external knowledge.
-          - Use the Source and Type metadata to understand context.
-          - If multiple results are relevant, synthesize them coherently.
-          - If information is insufficient, say so clearly.
-          - Be concise and direct; avoid meta commentary about the process.
-          - Results are ordered by relevance, with rank 1 being most relevant.
-        role: system
-      - content: Are there any class labels beyond the 11 listed for DocLayNet?
-        role: user
-      - content: |-
-          <think>
-          Need to search.
-          </think>
-        role: assistant
-        tool_calls:
-        - function:
-            arguments: '{"limit":5,"query":"DocLayNet class labels beyond the 11 listed"}'
-            name: search_and_answer
-          id: call_2dyiicy1
-          type: function
-      - content: |-
-          [f9612804-c5d1-4614-a86a-bff495f5ebf0] [rank 1 of 1]
-          Source: "DocLayNet Class Labels"
-          Type: list_item
-          Content:
-          DocLayNet defines 11 distinct class labels for document layout analysis:
-
-          Caption - Text describing figures or tables
-
-          Footnote - Notes at the bottom of pages
-
-          Formula - Mathematical expressions
-
-          List-item - Items in bulleted or numbered lists
-
-          Page-footer - Footer content on pages
-
-          Page-header - Header content on pages
-
-          Picture - Images and diagrams
-
-          Section-header - Headings for document sections
-
-          Table - Tabular data
-        role: tool
-        tool_call_id: call_2dyiicy1
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: Search the knowledge base for relevant documents.
-          name: search_and_answer
-          parameters:
-            additionalProperties: false
-            properties:
-              limit:
-                anyOf:
-                - type: integer
-                - type: 'null'
-                default: null
-              query:
-                type: string
-            required:
-            - query
-            type: object
-        type: function
-      - function:
-          description: Answer to a search query with chunk references.
-          name: final_result
-          parameters:
-            additionalProperties: false
-            properties:
-              answer:
-                description: The answer to the question
-                type: string
-              cited_chunks:
-                description: IDs of chunks used to form the answer
-                items:
-                  type: string
-                type: array
-              confidence:
-                default: 1.0
-                description: Confidence score for this answer (0-1)
-                maximum: 1.0
-                minimum: 0.0
-                type: number
-              query:
-                description: The question that was answered
-                type: string
-            required:
-            - query
-            - answer
-            type: object
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '603'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: stop
-        index: 0
-        message:
-          content: "query: Are there any class labels beyond the 11 listed for DocLayNet?  \nanswer: No, DocLayNet defines
-            exactly those 11 class labels (Caption, Footnote, Formula, List‑item, Page‑footer, Page‑header, Picture, Section‑header,
-            Table).  \ncited_chunks: [\"f9612804-c5d1-4614-a86a-bff495f5ebf0\"]  \nconfidence: 1.0"
-          role: assistant
-      created: 1768998216
-      id: chatcmpl-730
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 100
-        prompt_tokens: 821
-        total_tokens: 921
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '4200'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          You are a search and question-answering specialist.
-
-          Process:
-          1. Call search_and_answer with relevant keywords from the question.
-          2. Review the results ordered by relevance.
-          3. If needed, perform follow-up searches with different keywords (max 3 total).
-          4. Provide a concise answer based strictly on the retrieved content.
-
-          The search tool returns results like:
-          [9bde5847-44c9-400a-8997-0e6b65babf92] [rank 1 of 5]
-          Source: "Document Title" > Section > Subsection
-          Type: paragraph
-          Content:
-          The actual text content here...
-
-          [d5a63c82-cb40-439f-9b2e-de7d177829b7] [rank 2 of 5]
-          Source: "Another Document"
-          Type: table
-          Content:
-          | Column 1 | Column 2 |
-          ...
-
-          Each result includes:
-          - chunk_id in brackets and rank position (rank 1 = most relevant)
-          - Source: document title and section hierarchy (when available)
-          - Type: content type like paragraph, table, code, list_item (when available)
-          - Content: the actual text
-
-          Output format:
-          - query: Echo the question you are answering
-          - answer: Your concise answer based on the retrieved content
-          - cited_chunks: List of plain strings containing only the chunk UUIDs (not objects)
-          - confidence: A score from 0.0 to 1.0 indicating answer confidence
-
-          IMPORTANT: Use the EXACT, COMPLETE chunk ID (full UUID). Do NOT truncate IDs.
-
-          Guidelines:
-          - Base answers strictly on retrieved content - do not use external knowledge.
-          - Use the Source and Type metadata to understand context.
-          - If multiple results are relevant, synthesize them coherently.
-          - If information is insufficient, say so clearly.
-          - Be concise and direct; avoid meta commentary about the process.
-          - Results are ordered by relevance, with rank 1 being most relevant.
-        role: system
-      - content: Are there any class labels beyond the 11 listed for DocLayNet?
-        role: user
-      - content: |-
-          <think>
-          Need to search.
-          </think>
-        role: assistant
-        tool_calls:
-        - function:
-            arguments: '{"limit":5,"query":"DocLayNet class labels beyond the 11 listed"}'
-            name: search_and_answer
-          id: call_2dyiicy1
-          type: function
-      - content: |-
-          [f9612804-c5d1-4614-a86a-bff495f5ebf0] [rank 1 of 1]
-          Source: "DocLayNet Class Labels"
-          Type: list_item
-          Content:
-          DocLayNet defines 11 distinct class labels for document layout analysis:
-
-          Caption - Text describing figures or tables
-
-          Footnote - Notes at the bottom of pages
-
-          Formula - Mathematical expressions
-
-          List-item - Items in bulleted or numbered lists
-
-          Page-footer - Footer content on pages
-
-          Page-header - Header content on pages
-
-          Picture - Images and diagrams
-
-          Section-header - Headings for document sections
-
-          Table - Tabular data
-        role: tool
-        tool_call_id: call_2dyiicy1
-      - content: "query: Are there any class labels beyond the 11 listed for DocLayNet?  \nanswer: No, DocLayNet defines exactly
-          those 11 class labels (Caption, Footnote, Formula, List‑item, Page‑footer, Page‑header, Picture, Section‑header,
-          Table).  \ncited_chunks: [\"f9612804-c5d1-4614-a86a-bff495f5ebf0\"]  \nconfidence: 1.0"
-        role: assistant
-      - content: |-
-          Validation feedback:
-          Please include your response in a tool call.
-
-          Fix the errors and try again.
-        role: user
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: Search the knowledge base for relevant documents.
-          name: search_and_answer
-          parameters:
-            additionalProperties: false
-            properties:
-              limit:
-                anyOf:
-                - type: integer
-                - type: 'null'
-                default: null
-              query:
-                type: string
-            required:
-            - query
-            type: object
-        type: function
-      - function:
-          description: Answer to a search query with chunk references.
-          name: final_result
-          parameters:
-            additionalProperties: false
-            properties:
-              answer:
-                description: The answer to the question
-                type: string
-              cited_chunks:
-                description: IDs of chunks used to form the answer
-                items:
-                  type: string
-                type: array
-              confidence:
-                default: 1.0
-                description: Confidence score for this answer (0-1)
-                maximum: 1.0
-                minimum: 0.0
-                type: number
-              query:
-                description: The question that was answered
-                type: string
-            required:
-            - query
-            - answer
-            type: object
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '795'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: tool_calls
-        index: 0
-        message:
-          content: ''
-          reasoning: Need to wrap in function call final_result.
-          role: assistant
-          tool_calls:
-          - function:
-              arguments: '{"answer":"No, DocLayNet defines exactly those 11 class labels (Caption, Footnote, Formula, List‑item,
-                Page‑footer, Page‑header, Picture, Section‑header, Table).","cited_chunks":["f9612804-c5d1-4614-a86a-bff495f5ebf0"],"confidence":1,"query":"Are
-                there any class labels beyond the 11 listed for DocLayNet?"}'
-              name: final_result
-            id: call_qbsqj3yh
-            index: 0
-            type: function
-      created: 1768998218
-      id: chatcmpl-646
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 124
-        prompt_tokens: 946
-        total_tokens: 1070
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '3852'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          Generate a direct, conversational answer
-          to the question based on the gathered evidence.
-
-          Output:
-          - answer: Direct, comprehensive answer with a natural, helpful tone.
-            Write the actual answer, not a description of what you found.
-            Use as many sentences as needed to fully address the question.
-          - confidence: Score from 0.0 to 1.0 indicating answer quality.
-
-          Guidelines:
-          - Base your answer solely on the evidence provided in the context.
-          - If a <background> section is provided, use it to frame your answer appropriately.
-          - Be thorough - include all relevant information from the evidence.
-          - Use formatting (bullet points, numbered lists) when it improves clarity.
-          - Do NOT use meta-commentary like "Based on the research..." or "The evidence shows..."
-            Instead, directly state the information.
-          - If the evidence is incomplete, acknowledge limitations briefly.
-        role: system
-      - content: |-
-          Answer the question based on the gathered evidence.
-
-          <context>
-            <question>What class labels are defined in the dataset?</question>
-            <prior_answers>
-              <item>
-                <question>What are the 11 class labels in DocLayNet?</question>
-                <answer>The 11 class labels are: Caption, Footnote, Formula, List-item, Page-footer, Page-header, Picture, Section-header, Table, Text, and Title.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>How was the annotation process organized?</question>
-                <answer>The annotation was organized into 4 phases.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What data sources were used?</question>
-                <answer>Sources include arXiv and government offices.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What is the agreement metric?</question>
-                <answer>The mAP metric was used.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>What is machine learning?</question>
-                <answer>A field of AI.</answer>
-                <confidence>0.9</confidence>
-                <source>null</source>
-              </item>
-              <item>
-                <question>Are there any class labels beyond the 11 listed for DocLayNet?</question>
-                <answer>No, DocLayNet defines exactly those 11 class labels (Caption, Footnote, Formula, List‑item, Page‑footer, Page‑header, Picture, Section‑header, Table).</answer>
-                <confidence>1.0</confidence>
-                <source>DocLayNet Class Labels</source>
-              </item>
-            </prior_answers>
-          </context>
-        role: user
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: Conversational answer for chat context.
-          name: final_result
-          parameters:
-            $defs:
-              Citation:
-                additionalProperties: false
-                description: Resolved citation with full metadata for display/visual grounding.
-                properties:
-                  chunk_id:
-                    type: string
-                  content:
-                    type: string
-                  document_id:
-                    type: string
-                  document_title:
-                    anyOf:
-                    - type: string
-                    - type: 'null'
-                    default: null
-                  document_uri:
-                    type: string
-                  headings:
-                    anyOf:
-                    - items:
-                        type: string
-                      type: array
-                    - type: 'null'
-                    default: null
-                  page_numbers:
-                    items:
-                      type: integer
-                    type: array
-                required:
-                - document_id
-                - chunk_id
-                - document_uri
-                - content
-                type: object
-            additionalProperties: false
-            properties:
-              answer:
-                description: Direct answer to the question
-                type: string
-              citations:
-                description: Citations supporting the answer
-                items:
-                  $ref: '#/$defs/Citation'
-                type: array
-              confidence:
-                default: 1.0
-                description: Confidence score (0-1)
-                maximum: 1.0
-                minimum: 0.0
-                type: number
-            required:
-            - answer
-            type: object
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '809'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: tool_calls
-        index: 0
-        message:
-          content: ''
-          reasoning: 'Need answer question: What class labels are defined in the dataset? From prior answers, list all 11.
-            Provide confidence.'
-          role: assistant
-          tool_calls:
-          - function:
-              arguments: '{"answer":"The dataset contains 11 distinct class labels:\n\n1. Caption\n2. Footnote\n3. Formula\n4.
-                List‑item\n5. Page‑footer\n6. Page‑header\n7. Picture\n8. Section‑header\n9. Table\n10. Text\n11. Title","citations":[],"confidence":0.95}'
-              name: final_result
-            id: call_f6csb6me
-            index: 0
-            type: function
-      created: 1768998222
-      id: chatcmpl-608
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 121
-        prompt_tokens: 754
-        total_tokens: 875
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '4658'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      messages:
-      - content: |-
-          You are a helpful research assistant powered by haiku.rag, a knowledge base system.
-
-          You have access to a knowledge base of documents. Use your tools to search and answer questions.
-
-          CRITICAL RULES:
-          1. For greetings or casual chat: respond directly WITHOUT using any tools
-          2. For questions: Use the "ask" tool EXACTLY ONCE - it handles query expansion internally
-          3. For searches: Use the "search" tool EXACTLY ONCE - it handles multi-query expansion internally
-          4. NEVER call the same tool multiple times for a single user message
-          5. NEVER make up information - always use tools to get facts from the knowledge base
-
-          How to decide which tool to use:
-          - "get_document" - Use when the user references a SPECIFIC document by name, title, or URI (e.g., "summarize document X", "get the paper about Y", "fetch 2412.00566"). Retrieves the full document content.
-          - "ask" - Use for general questions about topics in the knowledge base when no specific document is named. It searches across all documents and returns answers with citations.
-          - "search" - Use when the user explicitly asks to search/find/explore documents. Call it ONCE. After calling search, copy the ENTIRE tool response to your output INCLUDING the content snippets. Do NOT shorten, summarize, or omit any part of the results.
-
-          IMPORTANT - When user mentions a document in search/ask:
-          - If user says "search in <doc>", "find in <doc>", "answer from <doc>", or "<topic> in <doc>":
-            - Extract the TOPIC as `query`/`question`
-            - Extract the DOCUMENT NAME as `document_name`
-          - Examples for search:
-            - "search for embeddings in the ML paper" → query="embeddings", document_name="ML paper"
-            - "find transformer architecture in 2412.00566" → query="transformer architecture", document_name="2412.00566"
-          - Examples for ask:
-            - "what does the ML paper say about embeddings?" → question="what are the embedding methods?", document_name="ML paper"
-            - "answer from 2412.00566 about model training" → question="how is the model trained?", document_name="2412.00566"
-
-          Be friendly and conversational. When you use the "ask" tool, summarize the key findings for the user.
-        role: system
-      - content: What class labels are defined in the dataset?
-        role: user
-      - content: |-
-          <think>
-          User asks about class labels defined in dataset. Use ask.
-          </think>
-        role: assistant
-        tool_calls:
-        - function:
-            arguments: '{"document_name":null,"question":"What class labels are defined in the dataset?"}'
-            name: ask
-          id: call_hcs4zx6t
-          type: function
-      - content: |-
-          The dataset contains 11 distinct class labels:
-
-          1. Caption
-          2. Footnote
-          3. Formula
-          4. List‑item
-          5. Page‑footer
-          6. Page‑header
-          7. Picture
-          8. Section‑header
-          9. Table
-          10. Text
-          11. Title
-
-          Sources: [1]
-        role: tool
-        tool_call_id: call_hcs4zx6t
-      model: gpt-oss
-      reasoning_effort: low
-      stream: false
-      tool_choice: auto
-      tools:
-      - function:
-          description: |-
-            Search the knowledge base for relevant documents.
-
-            Use this when you need to find documents or explore the knowledge base.
-            Results are displayed to the user - just list the titles found.
-          name: search
-          parameters:
-            additionalProperties: false
-            properties:
-              document_name:
-                anyOf:
-                - type: string
-                - type: 'null'
-                default: null
-                description: Optional document name/title to search within
-              limit:
-                anyOf:
-                - type: integer
-                - type: 'null'
-                default: null
-                description: 'Number of results to return (default: 5)'
-              query:
-                description: The search query (what to search for)
-                type: string
-            required:
-            - query
-            type: object
-        type: function
-      - function:
-          description: |-
-            Answer a specific question using the knowledge base.
-
-            Use this for direct questions that need a focused answer with citations.
-            Uses a research graph for planning, searching, and synthesis.
-          name: ask
-          parameters:
-            additionalProperties: false
-            properties:
-              document_name:
-                anyOf:
-                - type: string
-                - type: 'null'
-                default: null
-                description: Optional document name/title to search within (e.g., "tbmed593", "army manual")
-              question:
-                description: The question to answer
-                type: string
-            required:
-            - question
-            type: object
-        type: function
-      - function:
-          description: |-
-            Retrieve a specific document by title or URI.
-
-            Use this when the user wants to fetch/get/retrieve a specific document.
-          name: get_document
-          parameters:
-            additionalProperties: false
-            properties:
-              query:
-                description: The document title or URI to look up
-                type: string
-            required:
-            - query
-            type: object
-          strict: true
-        type: function
-    uri: http://localhost:11434/v1/chat/completions
-  response:
-    headers:
-      content-length:
-      - '629'
-      content-type:
-      - application/json
-    parsed_body:
-      choices:
-      - finish_reason: stop
-        index: 0
-        message:
-          content: "The dataset defines **11 distinct class labels**:\n\n1. Caption  \n2. Footnote  \n3. Formula  \n4. List‑item
-            \ \n5. Page‑footer  \n6. Page‑header  \n7. Picture  \n8. Section‑header  \n9. Table  \n10. Text  \n11. Title  \n\nThese
-            classes describe the different types of elements that can appear in the documents the dataset is built from."
-          role: assistant
-      created: 1768998225
-      id: chatcmpl-391
-      model: gpt-oss
-      object: chat.completion
-      system_fingerprint: fp_ollama
-      usage:
-        completion_tokens: 87
-        prompt_tokens: 977
-        total_tokens: 1064
-    status:
-      code: 200
-      message: OK
-version: 1
diff --git a/tests/cassettes/test_chat_state/test_rank_qa_history_returns_top_k.yaml b/tests/cassettes/test_chat_state/test_rank_qa_history_returns_top_k.yaml
deleted file mode 100644
index 4109736f..00000000
--- a/tests/cassettes/test_chat_state/test_rank_qa_history_returns_top_k.yaml
+++ /dev/null
@@ -1,126 +0,0 @@
-interactions:
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '123'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      encoding_format: base64
-      input:
-      - Which class label has the highest count in DocLayNet?
-      model: qwen3-embedding:4b
-    uri: http://localhost:11434/v1/embeddings
-  response:
-    headers:
-      content-type:
-      - application/json
-      transfer-encoding:
-      - chunked
-    parsed_body:
-      data:
-      - embedding: 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
-        index: 0
-        object: embedding
-      model: qwen3-embedding:4b
-      object: list
-      usage:
-        prompt_tokens: 14
-        total_tokens: 14
-    status:
-      code: 200
-      message: OK
-- request:
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate, zstd
-      connection:
-      - keep-alive
-      content-length:
-      - '933'
-      content-type:
-      - application/json
-      host:
-      - localhost:11434
-    method: POST
-    parsed_body:
-      encoding_format: base64
-      input:
-      - |-
-        Q: What are the 11 class labels in DocLayNet?
-        A: Caption, Footnote, Formula, List-item, Page-footer, Page-header, Picture, Section-header, Table, Text, and Title
-      - |-
-        Q: How was the annotation process organized?
-        A: The process had 4 phases with 40 dedicated annotators
-      - |-
-        Q: What data sources were used?
-        A: arXiv, government offices, company websites, financial reports and patents
-      - |-
-        Q: How were pages selected?
-        A: By selective subsampling with bias towards pages with figures or tables
-      - |-
-        Q: What is the inter-annotator agreement?
-        A: Computed as mAP@0.5-0.95 metric between pairwise annotations
-      - |-
-        Q: What is machine learning?
-        A: A field of AI that enables systems to learn from data
-      - |-
-        Q: How does neural network training work?
-        A: Through backpropagation and gradient descent
-      - |-
-        Q: What is deep learning?
-        A: A subset of ML using neural networks with many layers
-      model: qwen3-embedding:4b
-    uri: http://localhost:11434/v1/embeddings
-  response:
-    headers:
-      content-type:
-      - application/json
-      transfer-encoding:
-      - chunked
-    parsed_body:
-      data:
-      - embedding: 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
-        index: 0
-        object: embedding
-      - embedding: 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
-        index: 1
-        object: embedding
-      - embedding: 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
-        index: 2
-        object: embedding
-      - embedding: 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
-        index: 3
-        object: embedding
-      - embedding: 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
-        index: 4
-        object: embedding
-      - embedding: bWgjufDDtrwIvLu80Dmuuxt1gLq/Nl49Tb4DPUtSY71ChXo8rIROvQQtlD0lYgo8KvNjOiSRGL06kwA9Rf/gPAi7Fj2p2PQ8QNutPMlRgLt+FsS7JMFDPFCXv7xM1Bs9JrVPvB532bwtOYu8wKoNvSlUyLx9o0E8TmwoPcYrdryx1ws9lcHbu0z5vzusDf+8CmjMvNw9sLs4kHQ8vEq3vNmb4TyUy0u9jWZQPGX7aTwQj9q7NxRMvZGqdjsfDUo8B84GPCeTC71pbXU7/dasu/ytzjsQXMu8GNgUPYV41zyl4WM77wrau0++kDzH7VY8Fzt/uzfAcbrBrn68AH2pOi2G/7uTgsC8pXYtPIskCr1i2eo7IyAgPDCFhTsZGKw8lxYJvezeHTtDRzI6C8CHvNbVEruiO2o8ev0MPDoNszxyNsq8Yi5Zu0gGJjw6NAk9ZTmQPBhTtzzcVJS8caBvu2GTFL3TH6C7ipJHvLChQ7wahZs5EA6OvMWSNrvR+7g8z9tDvDU2lby/mBS8R+GeO3++GTp+VLA7xsHiPG2yULzR0qs8yDtPPEpHzLxAHCw7dO5dPHugOzt0mL071yUjPFHmODwgXUQ86GiQvDAX6rs0oxS9ms8mPUMfljw0Seq8DF0wvFO36jsGUoq535ZPvMmP4jtuUc075vO/O0/trbwdPdW8FH0/vJnXMbtvIwi8TWPQvC8tHLwVBB6791TKPKay9jux60o5h3BROtKFyztWoBu7MiJlPNjFhLv2u+Q8lPdHPEra3bzvqg27epXCOzNJbjx67oW81/zoO9JmJLxXgza6jfonPIXJ47qV2B88U8qNvO9XVzvyTTU8dPWiPDMRnjz1ilI8JOHtux71Qz2BUda6rjeXPEiTq7tUsmk8p6KsO9/6gbx2XPK6gM/hO9U2rbyeBC6808yxu5BNvzpIDnw8i1aTPAZFcLz+hbC85Hm+OkFX8zx31827s6h1PHXTqDzsD4Y80SRuu6kUCbw074C8heY/OwBhfDsWKo08g2cgukLWybu5H8m7StyRvOf/4DtXLPu85SXKPFmeyjwcZ+s8FueUuxfVAbxJm0M8FKWjO9GzRTsukJe7Sz6jvO+nqzzdDAC7v12yO7+g9zogso47ePrgPO06xTq1Ob05G1MxvGiO0LyUtoE8DGwPPYY1ALkVIRM8SjcavDYSpbxKm9a8CG03PHD3MTvx/mU7oCJCuymUGrxDSUA8Q85ZuxZLEDuxrI+82ivPuwZpVrwPLN28IOJrO1yTojzRhF47fo48vXBDObyjFPe6U9sUO0P2PbwNmcm8OHX3u878XryV4Hs6Z/hxu/k6eLu50hI8XehUvLku67v5IKE84uumu1ajLLwUTaC8etA5PE3txjzzT3i7VAgUvf1HErv8Ip48WGOnvFJMmjzIYMi73NMEvW0vvTw2YTi8mCsqvCIMK7tCawy7DFpqvJ21SDy6IQW9TwBvPIfyTbxQVje79DGYPH1tW7xjnM88yXJ2Ow1Csju75Rq9q4QvPVlLhT1I8gQ81U0MvfXPWjxnUyM87pcNvMdEkDyNr5Y7RRx5vTNmKj1doCE882+Xuykj0DuY9I68XtQjOgNMlrzxdhy9WRCAvHGCFLx3bis9i53wO10LbTySFO+7U4MPPPJUBbxt19w6y1cBPGPCkDy5b5k8Sg9BvcGdd7sJXmO6lySSvAy0G73dnIQ7G3VGvZVJ2ztTpV681OD/vOYFFTyk8iU8omrgPGt7ij3Dtqg77SEIPSLD0Twc2y+91EJdvEEpTTzbcMK7SRhEvDh81zyiiry7yLM5PG0UrDxK4BM9iiVquw4IIzuEhvG8ZcdxPJpwVTwDgh27CtOFvH50/rzsLly4QM0JvdSIRjvPdqC7JI1Hu51jnzubuKO8MLycvJMQzzyAC2I7xLByvO5eITvMuhG8b0zLPGlESr1I67G8VgKdvG4tqzsyrKs7CaYpPDMC77pvDs28QAXhO8h6pzuX6O+81CmdOcUIjrsz14C8E2aIPNAmQbyF2J08NOyjufpHl7znAwM8Bcr7uMq1oTsEYv67KSIrvEdWsTv57Ji7kI5uPELIITzOlUW8zwJ5O5DRkLxUCME8+4KOumxuGzuxNb68w7j+vGkzCLwjLP+85OslvM/xGL2qEgo6cjAWvHhbgjwNkPk8OCO1u7Pgk7zXVnE9BdZTPCYVETxKV5C8hPKaPOBE6LznHPK71wZIvfS0jDxSmQq60q6/uxbmZzlJV4q85PnlPMGyqjyViik8Ki26u8kkODwb6w88MApNPGc11Txobao8Y4QRPRV8ID1ZJhc8VveTPN41G70MkoK75j8mvKpWkbyu95I7jU4JPBdmSLvQVps8lf1TvSW8BLyJTDW8/YiTvC9AmLqP8+a82e+pvCMKt7tFc1E7lOpOvLJCzrwrogA96kmPOqY24zxtnsY7S3idPJ8oc73NEcq7NLXNup8bwDtIEXy83/ilvCZ9q7z3AgW8YyuEO3l5AryOX5e8I0/FOxUbKbpOcUs8P48Eu+iNhDwGUK+7GMuJPFt2mTolP/67PoQRu7l+cjwXT9Y84ikZPP7CRzuqG0g9KoKuO7ObyTzSVtI6L3VAvDy6ijxZzYe8dI4guoV1a7x7d7W6TRmOPLx1mTzsSAq8Hk8vO4NmYLu3ojI74u6hO5N/gDyxxVC7CI2FvCjVGj0gvVA5JfNPPCcIQ7uHKkk8YPCSPA9DFD3BZfe6AhTmvCB/BTquo947BaxFPAmJn7wxQNq7Tizxu7qba7xz+d28R8EIPFNAKTw/1OS88nkDPXmm6Dwf58C89kmTvJwJTDx0+EK8WMSJPJwL57sWr587X/ZbOufhAT2jW4K7SnNmvC7WUbxmQyq8cZ0bu6gCHzt/tRg8P3dlvM1ZbTwZE/y8CTX4O+sjd7yCvi+8oTZGvPXjDDsj+os8EaiQvPVTHzxgx3K8CkzZu9x/5Tz3ZEM8mw9lPbfnTTyUlUc8sEUwPCQNirwdnUG9z3ezPBdphrwfNpG8Gq8JPcb8+DvwQgO6GLBMvGFrtLw5uc46DYC9PEbpfbwxOBk9V8rIvNFf9rvw2J+80qLnO3QEojqNZPy8AW8XPcoUJb2r9MA8TakGPNgp97vRkCA8KqByOwB9t7y2SVy8q3wZO73ku7yp6IG8oIMSvNuq+7zaGoA8SWV9PBvGxrxvdqk6FMwoPMNINrw/Jqi8mYDKukZbHjxhuI480Z1hPINQzjzj3QW9nHEIPY2XkryTP7U6MLQ7Om8aJTs7Moa7KNX6u7pHpbwClpa8ip1pPL4ADD15i+C8eFYfvbh6tbxh27C8tH3Wux4aBL1udqE7gFqXO7BDJbzgOB+6jVu5Op877rvO6uI8rPN8vOg0tTui78K6EJwVvbY2zLs0DxI8cVi9uxCh3Tzds9G7Ff52u6AdlrpsFTM81yyfvKIIGrytbeg6QJqMuuAVoTsl8/Q7Ux4PPdaE8bzSpgo8omyguwrEZDtUUBg8eUqzPI32vbt/aoQ8tUK6O9E1pby4epS6Xz6bvJkQLLuwQTC8e43vvNAGkztJImw9mjkXu9T/zby8ciO8RcK/PEw4zDvtAP28Led4vJU96DxcC8M8uwp5O+dP1LxhjoU75daiu3zVJbt7W/O6aM08OSWRtTxKboG9EOdIO4jbuzvhhZO8xcadPKEyiDwuv4e65BG/PO1gmzsUaZG8uqfxOg2JirzOn7+8q5cGvTqQvbxsuPw8hlk8OvBIdjyDEIe8YLb4PNSwSLzDwQa8vwCtPIFvKjsDFIW87/8NvekgK72lWhi86nWPvPudsjyPB846OHwtvYz97LyqpOO8RWGfPC4KhryPEUm9840zvRGVTDx1M7e7g9iYO9/Bk7yCduS8TDTYPDDM1DwtCRe8E6JNPJBhiTvfGsK6v0ncOw7/urvTc/M7YTQdPF2+fLyjqWY81dX5vFmWWbuk9wg9sueNPENBaTy9+4S8db59vMI7iDuNdwW5wVaRvCATNTz9nVa8rH3tOVDjLjyrI2M8l8m2Ow8L1zvRi5I8fBOAvGkO2LxxI048IqVzuop8mryXUGK8LAagPG5fBj2BBvy5HD0eO78mR7waBqm4wj/dPEKZ+Dsbvwi8eo+XvLIQijnEY8C8G2GhvP87Aru3Bmk8fCNMvHEQwbybRz652eLPu4cJBTu0B3y8EXnkvJV6JL0UxDc8kr9EutovDz1PKAO81UKnO4SPnLxN/J88xYhFu77B0zzg9WQ8s1Qjvc/7ALuQyxk7imJRuhBwYrzwUla7eksyvELMpDucm747B531u6OQkDzwLyc9GfhtPDwDi7vV99e8/93Iu0jfczqsw8U7tizRPHCkO70qWss7BkpDPZ2M0Dwmkg08ZVTKPF9907y5Aps85NQcO4XHCzwmY5e8YS9fvKXqWTzrnv08Y7oqum3Yo7x0SSk86rHDPFzJCrz3YtM7UjaTPGTRobwtsAM9KcKFOsPSDj3JMWE7vaMyvFHK8bu4Nzq5JDDBugMsqjxwJpY78k/yukwjFz1uSzM9zqu8OxhtGT39Ukg93ObBO2EAiTyQ7es7jSpUPCPsUzz796w8TBFZvTzFnLwnCBK9Z3PTOoIqmLyiTzU8f+4vOg+q/jqMyDy8liz/OUD/R7zMDx89cPQyvKspebvVduk79qL7vLm8Dr1lxpW8lMZDPMELJTz/5HG61LgQPUSFyTyKkJw7ZP0NvaJK/Dxdl+C7yrf6OYpBsLxl6l28yjUfPVkCEr30LVg8nUonu+pSGL2B3DI8Jkveu3AAjDzVtKU6NV6OvCv2ujyxx2m8cW4PPcIEFj2Xptk8wTenvL0N9buhkwq7KXUFPa2/5LypNj28FPggvMsPo7zcOTG8SAGEPCiYoTxFu3M7qE5VvJX3fztqj+U8Sj2nvLjUkrxoe/e89/ELPRePNbxsfx29fAIKPJoFDL1Pe6O7T6GivDAiIrzt3MO8b/D+u2iPDTzL8EK8G2bkvLzvRrsXMQA99t5+u7b6n7x98+Q7zFd/vHTpnD0pGbg7701qPLHbr7uC2gs9St6EPEg19jz1zEe8tmEGu0ng5TpZ8KY8g0AWu5ZVXzvK3dO8uMmnPLqkqDzoV/w8oKHlPETRrbvpkSs8HY63O38FUrxfjrc8fuK7PJ+qs7vRYBK8Uc1PO4hSPb1oRL+7pkf8vKQCm7ymH7q8akE7PMbsdrwlqOU8Z/jHPLT0iLwi/RG9UkTuO1Z/Fzy0F8K7skiQunIa4ryaPgG9Wr4du7v9WryWdxG9nyRzPMk4nrtQTcS83Y6EuwpmcDwHfAw99IXsvICrDj2NrBw9SgsMPDFw2ztSzVo8zxr9PF+AB70ts528cVjEu2xO2jwwgSa8zZkxPGNFwrwFMGc7G6qau/oclTwMPT89ZjJSu9oijLzCnUq83u45vHKKmzyayJ88JLusPGl7Aj2T7BC9ZKlMPBltiDxI0zW8vKOHu36vxTyEGhS8rat3Owp8Bz0GlW465KiFvErbprzBDce6Ag62vDq77TtjT4u8vxXGPOD1SzpZxAW9mZV6O5djAj0WNlu6jtmLvOAxf7xPBAY8/X+BPI9pX7wtfbI81egduv8SDz3s0h68K6h/PFuqPrz5Rx480kLEO9miyDxIz6k84FB8u6AklzvHy6Q7axY8O1onILy/Yia92aLavFCofbztHYK7Ag72vLpOcz0bBte8dA+TvJ4ppLzQB/E8umK7PMa7Or2kAoi8pMbzuvtuzbvrI9w7aa/uvAF+BT1kuYE85Hqru3DG9TpvUL2861zjO0krXjxgvoU8ErbVvAAi4Ls8J+S7EOQGPHBbBrxkBa0820uXvDclZzvqhlE7tuGqvH+ITb0qdwM8r/R6PCRDCTyRThI8zoziuYrZ0zxQH8Q6RZVSPGiQCzzfvco8Cv3CvJwkzTx0eVC8nr/OPAgAMLxAynQ8Q6rtuwqtrLwEL2u814TRvKY8yTw7taQ857MzvMhdArya/uK7fE4uvMOLNrtm1YG6BvuBvLnE7zujlBo99BSsOzZIt7oSQUm8+7DnPDPyFTx7y648TgOZPGy1Yj0rAZI4jlTHvErQDrypFfi74HK6vMNgl7xlDD676SFpvLtWsbyuL/q7RiGJPNS8zzt+9qY6orjnPNqdxjo98w87jkEZuuRDTbtWwR08phGRPOoKPT2fetM5T5rIPPSghDu9hw89WazHvOC+ozvYQc68ofV/PLAQlzvbLgW8yT8XvIOBxLyQqMa8fDAlPNkHErwD4m08itEiPC6BBbz1b7I8t6nlvLHexDyySSa8jG8jPBtvBbyWzdC8BH+vPI6RbTz6KRE8PHnhuwwSoryLNCk9NaY2vC3msDwexzG9IIrpO7axjLyfxqG6jdZ/vZJAwbyeP0S8/dYpPMqURzz3zlu8LF64PM/0tjnwtfE75OSqvF/UbbuIgtu7AOtRPLc+PDxc7tq7T6GbuzISoTyGsWu8XRmBvJtFWDwvg/W88Jd7PK3iJr0aqvo7kgCVvLHGlDw4G4O8jwS9PCgCiTsl+Lm88RHaPJB007y7BtA8EIGhPEo8UbsuWTc9U+tSPV6h0jwV7QU9Pds3vCVmhrzDwXY8v7DCO0qB57tOMTC8yczDPMtMRzxKuD09dmRBPPddFroNU4k8CYgZPFBv3rx8Gq67rK6ZOyWC9Lt0a0e8CLAjPAf6mTtbvwQ9bjEUPeHgPDvzwy46ls5jvENuFrz/PJi7UeWZO0hBHztj5d476w28vBqJO7xi3Mu66l5SvFYsRz0WDgc9Fv+tvIkpLTyECuC7UPufuyabtrypd5C856wuO8Z9MDoQLWS7im97PHE7FLsqO8i7/7zouxvPuLw1PRu9Hhm4PDTomzxkpG68rIZiOtSEjzwEQe28Qy6suhHa/TrBwNm7M+jmugaMhryAAqm7vKPpvMd/kLzCh/48XyvRPE2bwDzIM4k6ZURqvAFw4TzhhrY3KMFiuz6Vgzz/ixI8BFkcvcnd9Do0soU7bhCmPOJJ3btAF3M6CwR3vFHEf71DJW28ClMau4SpFzyVzPM7aXMbPCGTg7wRuxy7AQPvuylu4LwY03C8JbUlOe+e2ryczRA5T0JAvF0aUjwDGmu8AIm4O2CY4jwyroq70t7NPHjw0ztJmig8dnWgOzLO3bwEyag6YN3SvIjvKztZREm8ockCvCfGejz/Tu68lPIou6qMIjpaOOU6oogIPCQ+mTtQwro6N9WFuvThsTvbgIK8zNgZPLGvxDsN/iK7S9YAPHZEibyPuJ87NcqAPJl5cbyz4VU814jwu+RPbzys/1Y8fZoEvKvDUTsgEdk8adJmPdsq0jzPdD67VO4tPLmTlTv3IeE6pcsLPNc2MzxuE5s8xBFKPQkAubdL7IY8KsbAvITKkTwgJU28TjqEPAZHGryZU4O8uyKCuyUMOLtvPN08fiCGvE13bjwXy4c8OF8vO+6h9TyxRDE8Lcu/u6JGLT1vfxY9GoQ0vHUvwLx4qi489QQoPEX/ML3Yt4U6My5zPH5V1DyfjYa89o4mvG6wWzxar/y6Dxv3PMtaj7x2gwO8KtIXvfY56TsHFuu8VXe3OpzCMbmIWhG9eIh3vHO32zycc1g8UBijO++LubvYKsK8p1JBvEpwx7zFYOI66tEPOR++bDyl/dY89VIgvNdxuTzqrDm6auW8vJOx1ryY+x09sELSPEc1CryAEGo8uUreOyz1xrwgXwc9wgeDPJynUTtllqC7fxtKPJcqtDzM0mk9FfPoPKWh0LyyTOy7Sf3lPO0ABDz2dS48hdQMPD/9Djze2QY8XEfZPCI7/LzbiAG8GAdnPKCABrx7PNy7p5GvvIBHTDxUFks8451jPJYYCLywEN07WbkDPa4TgzwwUc47JFw3vVESmrxrCcS7t3MaPdWmvzmK+c28xafCPLUUUL0oXWi7rNiku7mPxzyl7Pg8K5MAvWO1LDzRYw89pm/1PIhglDu9/mO8ZcsAvSF24juH4Nu8u+47u7hBiTwP26Y8djrNPK6xoDuACuU8yHUzOxbSIbs2xic6AvpoPFgMBD39Jjq9+f6iPFJmorshosi8itTCPCZzZ7yj/2G843LNu0ofuDzOfnC8doypu5bMvrtXPZG6J48rvAYxuDxoeHk82Aq2O6/p3TsJgSQ8jjMsPZMu0jtUBfW7jvl8PGHO6bukaic7UsxqvDfdATq4MBE9NWWPvEfCfTyWXq68+AJSPeiK2by0tem7A8vvO6PeyjmtuCM8ewbgu8sUmzsukIm8RO5Kux/XhzxVYX+8qtE3vGEACT20WMo89HWyPKTsnLsi7M48XchnOww6mzwWZle9wYRcu7fFFr1snfU7eeNxvMx0XjylD9S81WbpvLZN9rx86D+8Urk0PVzHBz0lSwC9s4vXO/jWOTxPmHG8OEgQu+Y/Dj3hdsM7hU1ivOB69Tyu+JG8tPB9vINZ+bu/BVe7P1gvvPCJjru+Utq8UKgQvTsGHTzRd+I8BLSiPMSvQ7towjm8Er46vRB6hbzIm4A8srhiPPqpSbnh/c68Dgm2Oid3qDvct9g7lNM8OZO72zrKV8W8geIJPNzLRrySrze8CwXtPDkrmDzeYGi81pN3O4YTl7xZugm72XMqPY0rBbwY3KM8XGgnPRIocDs65S+9jrAaveNhurvr9iI9QdchvBkpCr1zzVo8E2OnPE3MRjzcBqM8BD9BPAlAQ7wip+m84o65vLMBkbwrZQw93eTovDCwAL2I4WE7C8F7O38c7jwZxLq8ojPKO145XbwRGYC8E0GUPNXTwjxZL/U8AuwJPeg40bwgEGQ86lcAvMSrGDxU+6q7zpWrvF/uFrosZTo7RMEYPHaim7zOPRC9MroVPYZlXrycgq+8GxWpO7HH1bxHZIS8FeaCvKRL1zuhEY282243PCuGyDx5Uvg87UkzvbQVCzzl+Vq98DSSvEeYjDuT+Z28SojXPD8T3bwkqQo8SnHEPDBnVLwO9ZS8dIDYvP6dsjw+cEs7m0CDPMx0p7yiRNQ8nMCpujToHDzbMb07oMw7vEc3FLsfE1i861xuPB/xjzy3pY48jdifPLk//Lu28ZI7V/QLPaUsjLzzbYM7VoaHvB/Ohbvmn7S7EARavFeavDwk13u8oXhpvM95zzyzxJK7pCwTvHoR9zxUHVo8UIWkvOMsGruJY9S8X5XtvIKIkDyTZgE9epzSPL4lhjwHwly82adCuSH03Loecba7IKXAOiMPmTzfRIU7Z5pDvETNxrwjnQC9vgWwvEjribwdeH46e1KaPOFXODpZTNO8YbEzvYGKfDzrjUo8mgKZO6nohTv6nUk8QaIvu0Z+b7sra4o8WuFRO7r5VzwO7+i8DMgAvB0ye7xaI086WUQQPNaQzDvCUEk8768zvR0zpbwhirS8cbN9PDLcQrxU4qW8kwxxPGNvQTyVR+47SLE7vF+NZzzA2Ts76KFfPEBUAT0tEwU81uuJu8dZmzuxzKC8Np5yO9ojOLwM3wC8BQ+rPEd6+zdKX+k8y61tOzN0KDwWhYq8vU0UPAiHW7y4Zpi8U0oLOaUIgzvdj2y7QS2DvA/lRjx9jbS8EYUsPeOtSrznrG48WM9RPBjcd7wYdT47VR9OPD33JTxb7C09HErIuxvEubti6dK6iZbsukfYrLvnrgg9mVaVPHq4vryB7o08pBjgPEmVuLy1zTI9Po5FumINebxThfQ7nv0MveTd3bzFP0M5ju07vXCNdDwU7pe7TsQgvMByKbyzowi8RzfSvF/OPLxsQw29Bq68vGS5ozw5kCo8c/QOvIOJvrzJAoU816tKvKTmET2y47a7ZH8LPPJk5TqZkRU7B5HgvCD3hTzburC8A4CePPUUQbsGe4k7e/GrvDbCNrzwf3O8pVmZu/zh6jya2SM7Lu2TPLFHwbzs1aA5Le2AvGn60bwLfkk8oKekPMqol7pSe8u7yggEPWMn97xfTO28zBo0vEgZDDzaZpG8/8QdvH2TXDx1HLK85S3APIYM1LyFbGc87xROOxwlfTwH0J08X4+Ou8VJzTyylBw88y6lPHEoKjtftmI81IJbPFL96LyLcxW9WE8dvfFY3zxS6Wg8eivyO2Q0FjwvcNg7Ts3GvCdYIL2XuOO8cdv/PBYHgbwy3HG7JAn9vAgOMjuTOt66jEk1POdi/7swNgi9ZXNrvE6qMbunfu07TXKPO78TKzzatKO73amavND+Hz05EAs9vWZIPI3Y+zwIpui6SqKIvBaEnzxECis9MokvOnPLRLxkGdo88k6+umbHv7wMkxc88uhEO34z8Toj7i88ON9OO9WzcbyefaM8WQlfuyxviTthhxA7BIcMPQIQs7vO+y28R3ucPLnfPrwS8VI8zXRsu5kCrLrCfGI8sfaqO2mnrDwYdwk8TmAKOq4ZBb1iDLq8iTq8PN7ijLu94J68YXCzPMbNbDwlBxE8RIUAPJhdejtotV09Z1h1vD3Prrwlrq+8nd3OO9fcizxWJYu9NQGWPFzDXLvt2HA8+uAmvGePo7x20Xy8OwLCPIV2LDyw6GO74VO6O4emIrxpQWy8jhf6OyEVDb33++e8BOURvE530Tx5dv+84n0avMqWRjsKEHS8CzEDPd1b/js6Uza8k9tovFuJ2TwT/ms77XwCvSnWerziaRk93rSfvCBS0by1Jv689wZtO0O2aTw7Cjm8PsvAPI077bxANqG8nL4ZvZ7hHbsfrTm8/w5APEjlLb3iVEW7sjx5vF+/NrzkTR68x8i7vI5KqbyX+OY7+QEOvSq38TzkGNi6BFOGPAU4gbzJ/u48i2wJOtFVgjxNRBo7mETju3xplDxVG3U6At+fO3UpubsF8n08RT8FPYq/iLsgZIq8pjrRugcfxjwIbPU7gYhaO7P86zuw25E63/ISPc+ltLvZ6iW8te6+PPsrRryGjVI88MoEvCkkLjwxyJu6pQYXvCdbBLtVaPK7bf26PC810LyLlJ47ZqSKO5bZkjvepou9b8+kPFomsrpXP069IM0RvXuBPzuYkYk65SSxuVf6K7qi2vm7OEeovHFatTvkfre8Lp31PJeCj7xUf4I8eM8IvPNnEDsBttG8z4oPvIjImTstIbS8tmE+vJxYUDt/6C+9bxe/uwDGsrwigTS8jQXcvB0saLzDU/q8oEYNPN7L0bzNzVG8VqUePQ73MDvjDn+70qw8O28gB7thUGW9ovq6OuBQ9Lx4U/g4qvLKvB7RhbwpRk884w81vDGe2DzmYaq7vaequy1YujyIZja8Ki9iPNjRyLx0f5E7n3yLvJ3fa7unV/+8RrbrvELiIT1UcDo852iHPN6xA7zgmRI9oJU8O3tQobxo2cm7PWInvdrBqTuXgQI9KDRWvGiI0TzxdTU8mrxfuYDWkTxK1Xy8g7HLupZg07rFEu68s8BWOqV5Y7z/lJC8DsEEPbUxtjs69X87j02/u7UMjjw7Koa8kc3hu9zXkjwEirw8vimuPFamHrtalMO75Ok4PVPQzLzGeQ48KfBnvMS6lzp3blo8MoCnvLkggTwLGMK85UhyvHUHjrzmDXA5YdCdPLjnNr3J+YO8C+svvVuOrbulfqG8UvwcvcGQhLtenhK7Scr/vD23JzyXYyG9Vb8gPCiJv7xGb7k8+owDPFPrAzx7+xc9HPI7O47azbwDZxi8V0L4ukuGRz1fRje7eNFgu/r/zrsrjsk8Oq79u5PgorzO7Qu6Yf2hPKiuXLzVs2q7lBKiPJxf1Ly8Lge83AQQvLIAQrvGAHC8Z5Wlu0Z/H71ADi+9XHcJPTB78juT+lW9Xd9LvGFD47wYF7E8Z4AHvb1XZDzUNQ06UrmMvKAnCTqBfOO8NpCPPMPogTwkYBa8zOCWPFdJhrzidgY8cInIPJ11VDx5uy28+zaKu0mKPbzOpSw7WFivu+CdK73bZYk8oqCmO4QL7rzPPSo7/6hcuw3D1zxKrAm82mc8vA9Da7zsnx+7ojyXvLvt6bsqMze5ockxPN+r8bvmF0m88UAlvTfqAD3ukKc7pDaqPF3ivTyB8gA9/I0OPB2A47nlGeq5522/u0NAxjyCZKO87k5+vIPymTmeeiW8bGzKu8cjRLxxXfG6vXxdu8K6SzzdaSW83tMGu3ztoDxaUky8B4ITPODFCLwYFd+7kLpzu041wzwsN0a7D0MzPSZnnDyN83u88R3jO9l+27wykw+95KC9u/W4M7pV/8e73mGwPIMZH7wmcpk8jN5GPAcMursnsAW7n4OMvDddcDyn6Yw8cOSVOpsa0Tw+kvg6vp3bu8VkmztLnHm824QEPbwcID2yTKg8QvxSvIncvzyRoEk6JrnBvCt9b7zCqP67MCIWPLqomzwXz487C/6gvLyDfTzr6ww8OsLwvPrGk7wOosO6G/KQOrshOTwLD5Q7V7hmvPCNcjwcjfC7MIpdujTwqjy2frK6atiKu3nH2zvDQBU8RMfWvOPjKz0tYV06ETmVO4Cc1Tz8RoE8eYJRPVLPXDzeQaW8D64Wvaj+ITzDUB88TxxpPEWGRbuR2ha94H2YvPALQj2RWB48OC4qu41zsTulEmy6ecXWvJK/WbtVDG27hboCPOKYXry6Ery7EVVkvPm2JDxORxw94Ab/vMviIDxuNTq8qQikvHeusrpNIEQ80L/tu9k3yDvmwkY8cgRnPXmVp7xysBK80vS7PANO4DusFdG87whNu5lcUjw9HJ88VY+Fu7RDFz25j+08D0W3vKMptTw6NJO8JnRiPAUYSj1rGrE8uFK+O5ijlDpx21M9+qrsPMW0UDsaND278gJ1PM5bcTxsixg9GS4BPMJzfrw+J388ViZHvDOuhztupq68jkEKveYkMT2p28W76BEcPM88nLxBmT68rp94vMNAejtu4mI8j9kAO4k5ILyw7dE8OkXOu/Ae3Dv5/Aw8y4qcuzw/RDwAuwg9edlrPAncvbyrwMe71yi0vGYWxLyb9e686oXVPKsBl7yoYj885rHxPAkjAjrhM408b90zvJurhDuylfs7SdAmvOLuQzzpxn07J54EPTmcsjwy0KS8pVguO/7qqjxADuW6Q4WPPBomOTxGVvU7fZETugdC8TtOlj4752C9u6yN47siubW8T9JOvCF7ETz05Jc77AuRu42IuTx37vY8F8tJOyqJwzy4/ZK8OdGZPBtPcLzvDqu8YyEDvSXSXzzP5rw87AcvvcE7vTwP5BK7mhIgvOp6l7y2+7e8OWhpvO4QFzykqeq8wfUiPemauDvMe/e7Q7gzvM5fsjrOH+Y8osYTPLMTm7yPcly7FKaCOzzWNzzSTEc9mzy4vJWE5byJLKE8OMy9vPpauLz6S6o8ND/hu9SDEbssnDM8KQToO7tjFb0aVjG82FvavAIcqjzSjj68T9WMPN0jQLwOnzm7VMGUOqlXL7tOaDE9bJLmvMr/azxQCQA9i/f7us8/XLs+Pau8EptFu9+3SDxlhy28wFFLPMavBjvtjYc7X9+HvAR5NLxdJT089pvJu/GPUDy1r3E8dPOKvJ6+ybz0GLY8xdyKOx4QCjvi3aq82XeEvCZosrvgF7y7oK2AvNY6rbtpRnE8y8U7uwJEZjwEE4o7r44PvWy24jwfzxU7Uax/PG6l0jzaDAI8CZ+QPCzbcjwMqKo8Le2CO39TlDw6a5k8QP5qug==
-        index: 5
-        object: embedding
-      - embedding: 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
-        index: 6
-        object: embedding
-      - embedding: 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
-        index: 7
-        object: embedding
-      model: qwen3-embedding:4b
-      object: list
-      usage:
-        prompt_tokens: 210
-        total_tokens: 210
-    status:
-      code: 200
-      message: OK
-version: 1