diff --git a/BENCHMARKS.md b/BENCHMARKS.md
new file mode 100644
index 00000000..bc66cab3
--- /dev/null
+++ b/BENCHMARKS.md
@@ -0,0 +1,9 @@
+# `haiku.rag` benchmarks
+
+We use [repliqa](https://huggingface.co/datasets/ServiceNow/repliqa) for the evaluation of `haiku.rag`
+
+* Recall
+
+We load the `News Stories` from `repliqa_3` which is 1035 documents, using `tests/generate_benchmark_db.py`, using the `mxbai-embed-large` Ollama embeddings.
+
+Subsequently, we run a search over the `question` for each row of the dataset and check whether we match the document that answers the question. The recall obtained is ~0.75 for matching in the top result, raising to ~0.75 for the top 3 results.
diff --git a/tests/generate_benchmark_db.py b/tests/generate_benchmark_db.py
new file mode 100644
index 00000000..01ce7c20
--- /dev/null
+++ b/tests/generate_benchmark_db.py
@@ -0,0 +1,80 @@
+from pathlib import Path
+
+from datasets import Dataset, load_dataset
+from tqdm import tqdm
+
+from haiku.rag.client import HaikuRAG
+
+
+async def populate_db():
+    if (Path(__file__).parent / "benchmark.sqlite").exists():
+        print("Benchmark database already exists. Skipping creation.")
+        return
+
+    ds: Dataset = load_dataset("ServiceNow/repliqa")["repliqa_3"]  # type: ignore
+    corpus = ds.filter(lambda doc: doc["document_topic"] == "News Stories")
+
+    async with HaikuRAG(Path(__file__).parent / "benchmark.sqlite") as rag:
+        for i, doc in enumerate(tqdm(corpus)):
+            await rag.create_document(
+                content=doc["document_extracted"],  # type: ignore
+                uri=doc["document_id"],  # type: ignore
+            )
+
+
+async def run_match_benchmark():
+    ds: Dataset = load_dataset("ServiceNow/repliqa")["repliqa_3"]  # type: ignore
+    corpus = ds.filter(lambda doc: doc["document_topic"] == "News Stories")
+
+    correct_at_1 = 0
+    correct_at_2 = 0
+    correct_at_3 = 0
+    total_queries = 0
+
+    async with HaikuRAG(Path(__file__).parent / "benchmark.sqlite") as rag:
+        for i, doc in enumerate(tqdm(corpus)):
+            doc_id = doc["document_id"]  # type: ignore
+            matches = await rag.search(
+                query=doc["question"],  # type: ignore
+                limit=3,
+            )
+
+            total_queries += 1
+
+            # Check position of correct document in results
+            for position, (chunk, _) in enumerate(matches):
+                retrieved = await rag.get_document_by_id(chunk.document_id)
+                if retrieved and retrieved.uri == doc_id:
+                    if position == 0:  # First position
+                        correct_at_1 += 1
+                        correct_at_2 += 1
+                        correct_at_3 += 1
+                    elif position == 1:  # Second position
+                        correct_at_2 += 1
+                        correct_at_3 += 1
+                    elif position == 2:  # Third position
+                        correct_at_3 += 1
+                    break
+
+    # Calculate recall metrics
+    recall_at_1 = correct_at_1 / total_queries
+    recall_at_2 = correct_at_2 / total_queries
+    recall_at_3 = correct_at_3 / total_queries
+
+    print(f"Total queries: {total_queries}")
+    print(f"Recall@1: {recall_at_1:.4f}")
+    print(f"Recall@2: {recall_at_2:.4f}")
+    print(f"Recall@3: {recall_at_3:.4f}")
+
+    return {"recall@1": recall_at_1, "recall@2": recall_at_2, "recall@3": recall_at_3}
+
+
+async def main():
+    await populate_db()
+    await run_match_benchmark()
+
+
+if __name__ == "__main__":
+    import asyncio
+
+    asyncio.run(main())