realpython · realpython-bot · Sep 15, 2026
diff --git a/embeddings-and-vector-databases-with-chromadb/README.md b/embeddings-and-vector-databases-with-chromadb/README.md
@@ -1,11 +1,18 @@
-# Embeddings and Vector Databases With ChromaDB
+# ChromaDB: Embeddings and Vector Databases in Python
 
-Supporting code for the Real Python tutorial [Embeddings and Vector Databases With ChromaDB](https://realpython.com/chromadb-vector-database/). 
+Supporting code for the Real Python tutorial [ChromaDB: Embeddings and Vector Databases in Python](https://realpython.com/chromadb-vector-database/).
 
-To run the code in this tutorial, you should have `numpy`, `spacy`, `sentence-transformers`, `chromadb`, `polars`, `more-itertools`,  and `openai` installed in your environment. 
+The code was tested with Python 3.14 and the pinned versions in `requirements.txt`. You need Python 3.12 or later.
 
-You can install the dependencies manually, or by running:
+You can install the dependencies by running:
 
 ```
 (venv) $ python -m pip install -r requirements.txt
+(venv) $ python -m spacy download en_core_web_lg
+```
+
+To run the LLM examples, store your OpenAI API key in a `.env` file in this directory:
+
+```
+OPENAI_API_KEY="<your-api-key>"
 ```
diff --git a/embeddings-and-vector-databases-with-chromadb/car_data_etl.py b/embeddings-and-vector-databases-with-chromadb/car_data_etl.py
@@ -20,7 +20,7 @@ def prepare_car_reviews_data(
     }
 
     # Scan the car reviews dataset(s)
-    car_reviews = pl.scan_csv(data_path, dtypes=dtypes)
+    car_reviews = pl.scan_csv(data_path, schema_overrides=dtypes)
 
     # Extract the vehicle title and year as new columns
     # Filter on selected years
@@ -48,7 +48,7 @@ def prepare_car_reviews_data(
                 "Vehicle_Model",
             ]
         )
-        .sort(["Vehicle_Model", "Rating"])
+        .sort(["Vehicle_Model", "Rating"], maintain_order=True)
         .collect()
     )
 

diff --git a/embeddings-and-vector-databases-with-chromadb/chroma_utils.py b/embeddings-and-vector-databases-with-chromadb/chroma_utils.py
@@ -25,14 +25,15 @@ def build_chroma_collection(
     collection = chroma_client.create_collection(
         name=collection_name,
         embedding_function=embedding_func,
-        metadata={"hnsw:space": distance_func_name},
+        configuration={"hnsw": {"space": distance_func_name}},
     )
 
+    batch_size = chroma_client.get_max_batch_size()
     document_indices = list(range(len(documents)))
 
-    for batch in batched(document_indices, 166):
+    for batch in batched(document_indices, batch_size):
         start_idx = batch[0]
-        end_idx = batch[-1]
+        end_idx = batch[-1] + 1
 
         collection.add(
             ids=ids[start_idx:end_idx],

diff --git a/embeddings-and-vector-databases-with-chromadb/config.json b/embeddings-and-vector-databases-with-chromadb/config.json
diff --git a/embeddings-and-vector-databases-with-chromadb/create_car_review_collection.py b/embeddings-and-vector-databases-with-chromadb/create_car_review_collection.py
@@ -1,7 +1,6 @@
 import chromadb
 from car_data_etl import prepare_car_reviews_data
 from chroma_utils import build_chroma_collection
-from chromadb.utils import embedding_functions
 
 DATA_PATH = "data/archive/*"
 CHROMA_PATH = "car_review_embeddings"
@@ -20,12 +19,9 @@
 )
 
 client = chromadb.PersistentClient(CHROMA_PATH)
-embedding_func = embedding_functions.SentenceTransformerEmbeddingFunction(
-    model_name=EMBEDDING_FUNC_NAME
-)
-collection = client.get_collection(
-    name=COLLECTION_NAME, embedding_function=embedding_func
-)
+collection = client.get_collection(name=COLLECTION_NAME)
+
+print(collection.count())
 
 great_reviews = collection.query(
     query_texts=[

diff --git a/embeddings-and-vector-databases-with-chromadb/llm_car_review_context.py b/embeddings-and-vector-databases-with-chromadb/llm_car_review_context.py
@@ -1,41 +1,31 @@
-import json
 import os
 
 import chromadb
-import openai
-from chromadb.utils import embedding_functions
+from dotenv import load_dotenv
+from openai import OpenAI
 
 os.environ["TOKENIZERS_PARALLELISM"] = "false"
 
-DATA_PATH = "data/archive/*"
 CHROMA_PATH = "car_review_embeddings"
-EMBEDDING_FUNC_NAME = "multi-qa-MiniLM-L6-cos-v1"
 COLLECTION_NAME = "car_reviews"
+MODEL = "gpt-5.6-luna"
 
-with open("config.json", "r") as json_file:
-    config_data = json.load(json_file)
+load_dotenv()
 
-openai.api_key = config_data.get("openai-secret-key")
-
-client = chromadb.PersistentClient(CHROMA_PATH)
-embedding_func = embedding_functions.SentenceTransformerEmbeddingFunction(
-    model_name=EMBEDDING_FUNC_NAME
-)
-
-collection = client.get_collection(
-    name=COLLECTION_NAME, embedding_function=embedding_func
-)
+openai_client = OpenAI()
+chroma_client = chromadb.PersistentClient(CHROMA_PATH)
+collection = chroma_client.get_collection(name=COLLECTION_NAME)
 
 context = """
- You are a customer success employee at a large
-  car dealership. Use the following car reviews
-  to answer questions: {}
- """
+You are a customer success employee at a large
+ car dealership. Use the following car reviews
+ to answer questions: {}
+"""
 
 question = """
- What's the key to great customer satisfaction
-  based on detailed positive reviews?
- """
+What's the key to great customer satisfaction
+ based on detailed positive reviews?
+"""
 
 good_reviews = collection.query(
     query_texts=[question],
@@ -46,45 +36,25 @@
 
 reviews_str = ",".join(good_reviews["documents"][0])
 
-good_review_summaries = openai.ChatCompletion.create(
-    model="gpt-3.5-turbo",
-    messages=[
-        {"role": "system", "content": context.format(reviews_str)},
-        {"role": "user", "content": question},
-    ],
-    temperature=0,
-    n=1,
-)
-
-reviews_str = ",".join(good_reviews["documents"][0])
-
 print("Good reviews: ")
 print(reviews_str)
 print("###########################################")
 
-good_review_summaries = openai.ChatCompletion.create(
-    model="gpt-3.5-turbo",
-    messages=[
-        {"role": "system", "content": context.format(reviews_str)},
-        {"role": "user", "content": question},
-    ],
-    temperature=0,
-    n=1,
+good_review_summaries = openai_client.responses.create(
+    model=MODEL,
+    instructions=context.format(reviews_str),
+    input=question,
 )
 
 print("AI-Generated summary of good reviews: ")
-print(good_review_summaries["choices"][0]["message"]["content"])
+print(good_review_summaries.output_text)
 print("###########################################")
 
-
-context = """
-          You are a customer success employee at a large car dealership.
-          Use the following car reivews to answer questions: {}
-          """
 question = """
-            Which of these poor reviews has the worst implications about
-            our dealership? Explain why.
-            """
+Which of these poor reviews has the
+ worst implications about our dealership?
+ Explain why.
+"""
 
 poor_reviews = collection.query(
     query_texts=[question],
@@ -99,16 +69,12 @@
 print(poor_reviews["documents"][0][0])
 print("###########################################")
 
-poor_review_analysis = openai.ChatCompletion.create(
-    model="gpt-3.5-turbo",
-    messages=[
-        {"role": "system", "content": context.format(reviews_str)},
-        {"role": "user", "content": question},
-    ],
-    temperature=0,
-    n=1,
+poor_review_analysis = openai_client.responses.create(
+    model=MODEL,
+    instructions=context.format(reviews_str),
+    input=question,
 )
 
 print("AI-Generated summary of the single worst review: ")
-print(poor_review_analysis["choices"][0]["message"]["content"])
+print(poor_review_analysis.output_text)
 print("###########################################")
diff --git a/embeddings-and-vector-databases-with-chromadb/requirements.txt b/embeddings-and-vector-databases-with-chromadb/requirements.txt
@@ -1,92 +1,7 @@
-aiohttp==3.8.6
-aiosignal==1.3.1
-annotated-types==0.6.0
-anyio==3.7.1
-async-timeout==4.0.3
-attrs==23.1.0
-backoff==2.2.1
-bcrypt==4.0.1
-blis==0.7.11
-catalogue==2.0.10
-certifi==2023.7.22
-charset-normalizer==3.3.0
-chroma-hnswlib==0.7.3
-chromadb==0.4.14
-click==8.1.7
-cloudpathlib==0.16.0
-coloredlogs==15.0.1
-confection==0.1.3
-cymem==2.0.8
-fastapi==0.104.0
-filelock==3.12.4
-flatbuffers==23.5.26
-frozenlist==1.4.0
-fsspec==2023.9.2
-grpcio==1.59.0
-h11==0.14.0
-httptools==0.6.1
-huggingface-hub==0.17.3
-humanfriendly==10.0
-idna==3.4
-importlib-resources==6.1.0
-Jinja2==3.1.2
-joblib==1.3.2
-langcodes==3.3.0
-MarkupSafe==2.1.3
-monotonic==1.6
-more-itertools==10.1.0
-mpmath==1.3.0
-multidict==6.0.4
-murmurhash==1.0.10
-networkx==3.2
-nltk==3.8.1
-numpy==1.26.1
-onnxruntime==1.16.1
-openai==0.28.1
-overrides==7.4.0
-packaging==23.2
-Pillow==10.1.0
-polars==0.19.9
-posthog==3.0.2
-preshed==3.0.9
-protobuf==4.24.4
-pulsar-client==3.3.0
-pydantic==2.4.2
-pydantic_core==2.10.1
-PyPika==0.48.9
-python-dateutil==2.8.2
-python-dotenv==1.0.0
-PyYAML==6.0.1
-regex==2023.10.3
-requests==2.31.0
-safetensors==0.4.0
-scikit-learn==1.3.1
-scipy==1.11.3
-sentence-transformers==2.2.2
-sentencepiece==0.1.99
-six==1.16.0
-smart-open==6.4.0
-sniffio==1.3.0
-spacy==3.7.2
-spacy-legacy==3.0.12
-spacy-loggers==1.0.5
-srsly==2.4.8
-starlette==0.27.0
-sympy==1.12
-thinc==8.2.1
-threadpoolctl==3.2.0
-tokenizers==0.14.1
-torch==2.1.0
-torchvision==0.16.0
-tqdm==4.66.1
-transformers==4.34.1
-typer==0.9.0
-typing_extensions==4.8.0
-urllib3==2.0.7
-uvicorn==0.23.2
-uvloop==0.18.0
-wasabi==1.1.2
-watchfiles==0.21.0
-weasel==0.3.3
-websockets==11.0.3
-yarl==1.9.2
+chromadb==1.5.9
+numpy==2.5.3
+openai==3.14.0
+polars==1.44.2
+python-dotenv==1.2.3
+sentence-transformers==6.0.1
+spacy==3.8.16
diff --git a/embeddings-and-vector-databases-with-chromadb/word_vectors.py b/embeddings-and-vector-databases-with-chromadb/word_vectors.py
@@ -3,7 +3,7 @@
 import spacy
 
 # Load the medium-size English model
-nlp = spacy.load("en_core_web_md")
+nlp = spacy.load("en_core_web_lg")
 
 # Get the word vector for the word "dog"
 dog_embedding = nlp.vocab["dog"].vector