OpenAI AWS Lambda Layer (Open Source)

[eren]

For anyone looking to deploy scalable and cost-efficient micro services using AWS Lambda for your OpenAI project, check out my repo:

GitHub - erenyasarkurt/OpenAI-AWS-Lambda-Layer: OpenAI AWS Lambda Layer....

You can also use Lambda concurrency to implement rate-limiting required by OpenAI!

[jsonkha]

hello, I modified your code to use 3.10 but it doesn’t work. Why don’t you use an official aws amazon linux ?

[lpm0073]

Nicely done @eren! I’ve also published something somewhat similar. See FullStackWithLawrence/aws-openai.

This creates A REST API implementing each of the 30 example applications from the official OpenAI API Documentation. Implemented as a serverless microservice using AWS API Gateway, Lambda and the OpenAI Python Library. Leverages OpenAI’s suite of AI models, including GPT-3.5, GPT-4, DALL·E, Whisper, Embeddings, and Moderation.

[curt.kennedy]

Cool, but why wouldn’t I just create my own lambda layer from the OpenAI SDK?

Is this for people that don’t know how to create their own layers? It only takes a few minutes! What am I missing?

[tomsoderlund]

Same question here, but I get “Service Unavailable” messages when I try to call e.g. openai.audio.transcriptions.create with the OpenAI Node.js API from an AWS Lambda.

Do I need some kind of OpenAI Lambda layer if I’m just using the OpenAI Node.js API?

Update: my issue could be Usage Limit related, investigating…

[curt.kennedy]

If it helps, when the Whisper API was first announced, I had trouble calling it. But you can call it directly without the SDK.

Here is a post where the Python version of that code lives, and this is what I run from Lambda.

Introducing ChatGPT and Whisper APIs

If you want to get Whisper transcriptions in python and not deal with the overhead of the OpenAI python bindings, here is the python code I came up with: import requests import os ApiKey = os.environ.get("ApiKey") headers = { 'Authorization': f'Bearer {ApiKey}' } files = { 'file': open('/local/path/to/your/file/audio.wav', 'rb'), 'model': (None, 'whisper-1'), } response = requests.post('https://api.openai.com/v1/audio/transcriptions', headers=headers, files=files) print(respons…

[tallamsahil5]

Hey, I see that the layer was built using the OpenAI version 0.27.4. I tried updating it by doin pip install openai==1.2.1, but when creating the layer in lambda and testing the code I am getting the error:

errorMessage": “Unable to import module ‘lambda_function’: No module named ‘openai’”

Anyone mind helping out?