diff --git a/docs/sdk/quickstart.md b/docs/sdk/quickstart.md
deleted file mode 100644
index fd4e069..0000000
--- a/docs/sdk/quickstart.md
+++ /dev/null
@@ -1,67 +0,0 @@
-# Quickstart
-
-!!! note "Prerequisites"
-    - Python 3.11+
-    - Install the package: `pip install openagent-eval`
-    - A compatible LLM provider (e.g., OpenAI) and any required credentials are available as environment variables.
-
-The following example shows the smallest possible workflow: create a configuration, run the evaluation engine on a tiny in‑memory dataset, and print the report summary.
-
-```python
-"""Minimal OpenAgent Eval example.
-Runs an async engine with a single question and prints the summary.
-"""
-
-import asyncio
-from pathlib import Path
-
-from openagent_eval.config.models import Config, LLMConfig, RetrieverConfig, MetricsConfig, DatasetConfig
-from openagent_eval.core.engine import Engine
-
-# 1️⃣  Build a configuration
-config = Config(
-    dataset=DatasetConfig(path="data/sample.json", limit=1),
-    llm=LLMConfig(provider="openai", model="gpt-4o-mini", temperature=0.0),
-    retriever=RetrieverConfig(provider="mock"),  # simple mock retriever
-    metrics=MetricsConfig(
-        retrieval=[],
-        generation=["faithfulness"],
-        performance=[],
-        cost=[],
-    ),
-)
-
-# 2️⃣  Create the engine
-engine = Engine(config)
-
-# 3️⃣  Define a tiny dataset inline (overrides the path above)
-#    The engine expects a list of dicts matching `DatasetItemModel`.
-dataset = [
-    {
-        "question": "What is the capital of France?",
-        "ground_truth": "Paris",
-        "metadata": {},
-    }
-]
-
-# 4️⃣  Run the evaluation (async)
-async def main():
-    report = await engine.run(dataset)
-    # The report contains a handy `summary` dict – print the most useful bits.
-    print("✅ Evaluation complete")
-    print("Total items:", report.summary["total_items"])
-    print("Successful:", report.summary["successful_evaluations"])
-    print("Metrics summary:", report.summary.get("metrics_summary", {}))
-    # Optionally persist the report using ReportManager (see next docs page).
-
-asyncio.run(main())
-```
-
-**What you just saw**
-- `Config` ties together dataset, LLM, retriever, and metrics.
-- `Engine` orchestrates provider construction, pipeline execution, and report generation.
-- The example uses the built‑in `mock` retriever so no external service is needed.
-
-Next, dive deeper into the mental model that powers this SDK.
-
-[Next: Concepts →](concepts.md)
diff --git a/openagent_eval/__init__.py b/openagent_eval/__init__.py
index 906abb7..c2aba67 100644
--- a/openagent_eval/__init__.py
+++ b/openagent_eval/__init__.py
@@ -1,4 +1,4 @@
 """OpenAgent Eval - Open-source CLI framework for evaluating RAG systems and AI Agents."""
 
-__version__ = "0.4.5"
+__version__ = "0.4.6"
 __author__ = "OpenAgentHQ"
diff --git a/pyproject.toml b/pyproject.toml
index 480108e..df022b6 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "openagent-eval"
-version = "0.4.5"
+version = "0.4.6"
 description = "Open-source CLI framework for evaluating RAG systems and AI Agents"
 readme = "README.md"
 license = {text = "Apache-2.0"}