aws-lambda-layer-openai

This repository provides steps to create a custom AWS Lambda layer for the OpenAI library, including the latest version .zip file for direct deployment.

Step 1: Create a virtual environment

Create a Python virtual environment and activate it.

# Create the aws-lambda-layer directory and navigate into it
mkdir aws-lambda-layer
cd aws-lambda-layer

# Create a virtual environment named 'myenv' with Python 3.10
python3.10 -m venv myenv

# Activate the virtual environment
source ./myenv/bin/activate

Step 2: List your dependencies in requirements.txt

The OpenAi Python library depends on quite a few other libraries. Make sure to list all of them in the requirements.txt file along with their desired versions. This ensures that all their built distributions are installed with respect to your targeted platform. Place your requirements.txt file inside the above directory aws-lambda-layer

Step 3: Install built distributions (wheels)

To download a wheel that’s compatible with Lambda, you use the pip --platform option.

If your Lambda function uses the x86_64 instruction set architecture, run the following pip install command to install a compatible wheel in your package directory. Replace --python 3.x with the version of the Python runtime you are using.

pip install \
--platform manylinux2014_x86_64 \
--target=package \
--implementation cp \
--python-version 3.10 \
--only-binary=:all: --upgrade \
-r requirements.txt

If your Lambda function uses the arm64 instruction set architecture, run the following pip install command to install a compatible wheel in your package directory. Replace --python 3.x with the version of the Python runtime you are using.

pip install \
--platform manylinux2014_aarch64 \
--target=package \
--implementation cp \
--python-version 3.10 \
--only-binary=:all: --upgrade \
-r requirements.txt

Step 4: Deactivate the environment

deactivate

Step 5: Create a .zip file for the downloaded packages

# Create the 'python' folder
mkdir python

# Copy all contents from 'package' to 'python'
cp -r package/* python/

# Zip the 'python' folder
zip -r <zip-file-name>.zip python

Step 6: Upload the .zip file as custom layer over aws console

1. Navigate to Lambda -> Layers
2. Press Create Layer
3. Write your layer name, description
4. Upload .zip file
5. Choose compatible architectures (x86_64, arm64)
6. Choose compatible runtimes (Python3.10, Python3.11, Python3.12)
7. Press Create