diff --git a/examples/basic/vector_ollama.py b/examples/basic/vector_ollama.py
new file mode 100644
index 0000000..c7b2399
--- /dev/null
+++ b/examples/basic/vector_ollama.py
@@ -0,0 +1,95 @@
+"""
+Use CrateDB Vector Search with embeddings computed by a local Ollama server.
+
+Ollama runs the embedding model in its own process and speaks HTTP, so this
+program needs no account, no API key, and no machine learning stack of its
+own. `vector_ollama.py` and `vector_openai.py` are otherwise the same program.
+
+- https://ollama.com/library/nomic-embed-text
+- https://python.langchain.com/docs/integrations/text_embedding/ollama/
+
+As input data, the example uses the canonical `state_of_the_union.txt`.
+
+Synopsis::
+
+    # Install prerequisites.
+    pip install --upgrade langchain-cratedb langchain-ollama langchain-text-splitters
+
+    # Start database.
+    docker run --rm -it --publish=4200:4200 crate/crate:nightly
+
+    # Serve the embedding model. `nomic-embed-text` produces 768 dimensions,
+    # well within the 2048 a CrateDB FLOAT_VECTOR column accepts.
+    ollama serve
+    ollama pull nomic-embed-text
+
+    # Optionally set environment variables to configure the Ollama and CrateDB
+    # endpoints.
+    export OLLAMA_BASE_URL="http://localhost:11434"
+    export CRATEDB_SQLALCHEMY_URL="crate://crate@localhost/?schema=doc"
+
+    # Run program.
+    python examples/basic/vector_ollama.py
+"""  # noqa: E501
+# /// script
+# requires-python = ">=3.10"
+# dependencies = [
+#   "langchain-cratedb",
+#   "langchain-ollama",
+#   "langchain-text-splitters",
+# ]
+# ///
+
+import os
+import typing as t
+
+import requests
+from langchain_core.documents import Document
+from langchain_ollama import OllamaEmbeddings
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+
+from langchain_cratedb import CrateDBVectorStore
+
+CRATEDB_SQLALCHEMY_URL = os.environ.get(
+    "CRATEDB_SQLALCHEMY_URL", "crate://crate@localhost/?schema=testdrive"
+)
+OLLAMA_BASE_URL = os.environ.get("OLLAMA_BASE_URL", "http://localhost:11434")
+EMBEDDING_MODEL = "nomic-embed-text"
+
+
+def get_documents() -> t.List[Document]:
+    """
+    Acquire data, return as LangChain documents.
+    """
+
+    # Define text splitter.
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
+
+    # Load a document, and split it into chunks.
+    url = "https://github.com/langchain-ai/langchain/raw/v0.0.325/docs/docs/modules/state_of_the_union.txt"
+    text = requests.get(url, timeout=10).text
+    return text_splitter.create_documents([text])
+
+
+def main() -> None:
+    # Set up the embedding model.
+    embeddings = OllamaEmbeddings(model=EMBEDDING_MODEL, base_url=OLLAMA_BASE_URL)
+
+    # Acquire documents.
+    documents = get_documents()
+
+    # Embed each chunk, and load them into the vector store.
+    vector_store = CrateDBVectorStore.from_documents(
+        documents=documents,
+        embedding=embeddings,
+        connection=CRATEDB_SQLALCHEMY_URL,
+    )
+
+    # Invoke a query, and display the first result.
+    query = "What did the president say about Ketanji Brown Jackson"
+    docs = vector_store.similarity_search(query)
+    print(docs[0].page_content)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/examples/basic/vector_search.py b/examples/basic/vector_openai.py
similarity index 94%
rename from examples/basic/vector_search.py
rename to examples/basic/vector_openai.py
index 4b5e003..edd6090 100644
--- a/examples/basic/vector_search.py
+++ b/examples/basic/vector_openai.py
@@ -1,6 +1,8 @@
 """
 Use CrateDB Vector Search with OpenAI embeddings.
 
+For the same program without an account or an API key, see `vector_ollama.py`.
+
 As input data, the example uses the canonical `state_of_the_union.txt`.
 
 Synopsis::
@@ -17,7 +19,7 @@
     export CRATEDB_SQLALCHEMY_URL="crate://crate@localhost/?schema=doc"
 
     # Run program.
-    python examples/basic/vector_search.py
+    python examples/basic/vector_openai.py
 """  # noqa: E501
 # /// script
 # requires-python = ">=3.9"
diff --git a/pyproject.toml b/pyproject.toml
index 40cafdf..2f1ba7c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -147,6 +147,9 @@ select = [
 omit = [
     "langchain_cratedb/retrievers.py",
     "tests/*",
+    # Runs against a local Ollama server, which CI does not provide, so the
+    # suite reaches its import and skips the rest.
+    "examples/basic/vector_ollama.py",
 ]
 
 [tool.pytest.ini_options]
@@ -180,6 +183,7 @@ optional = true
 [tool.poetry.group.dev.dependencies]
 
 [tool.poetry.group.test.dependencies]
+langchain-ollama = "<2"
 langchain-openai = "<1.3"
 langchain-tests = "==1.1.7"
 notebook = "<7.6"
diff --git a/tests/test_examples.py b/tests/test_examples.py
index 404934e..480fa1d 100644
--- a/tests/test_examples.py
+++ b/tests/test_examples.py
@@ -35,7 +35,8 @@ def test_file(run_file: t.Callable, file: Path) -> None:
     """
     Execute Python code, one test case per .py file.
 
-    Skip test cases that trip when no OpenAI API key is configured.
+    Skip test cases that need a service this machine does not provide: an
+    OpenAI API key, or a reachable Ollama server.
     """
     if file.name in SKIP_FILES:
         raise pytest.skip(f"FIXME: Skipping file: {file.name}")
@@ -47,3 +48,7 @@ def test_file(run_file: t.Callable, file: Path) -> None:
         raise pytest.skip(
             "Skipping test because `OPENAI_API_KEY` is not defined"
         ) from ex
+    except ConnectionError as ex:
+        if "Failed to connect to Ollama" not in str(ex):
+            raise
+        raise pytest.skip("Skipping test because Ollama is not reachable") from ex