Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature]: aiohttp migration - 10-100x Higher RPS Master ticket #7544

Open
1 of 9 tasks
ishaan-jaff opened this issue Jan 4, 2025 · 2 comments
Open
1 of 9 tasks

[Feature]: aiohttp migration - 10-100x Higher RPS Master ticket #7544

ishaan-jaff opened this issue Jan 4, 2025 · 2 comments
Labels
enhancement New feature or request

Comments

@ishaan-jaff
Copy link
Contributor

ishaan-jaff commented Jan 4, 2025

The Feature

Creating this ticket to breakdown the larger problem

  • Ensure litellm proxy can hit 2k RPS, at 140ms median latency with 4 instances
  • Use provider config request / response transformations in BaseLLMAIOHTTPHandler
  • Add streaming support for aiohttp_openai/
  • Add testing for using aiohttp_openai with litellm router
  • Migrate BaseLLMHTTPHandler to use BaseLLMAIOHTTPHandler
  • Migrate openai_like to BaseLLMAIOHTTPHandler
  • Migrate openai, azure to BaseLLMAIOHTTPHandler
  • Migrate huggingface/ to BaseLLMAIOHTTPHandler
  • Migrate anthropic/ to BaseLLMAIOHTTPHandler

Motivation, pitch

Are you a ML Ops Team?

No

Twitter / LinkedIn details

No response

@mirodrr2
Copy link

Does this need to be done for bedrock? Or is bedrock support already implemented in an async way?

@ishaan-jaff
Copy link
Contributor Author

ishaan-jaff commented Jan 28, 2025

bedrock is already async @mirodrr2 - but all providers are going to be migrated to aiohttp. So yes needs to be done for bedrock too. Can you help @mirodrr2 ?

It's a bit involed so happy to get on a call and discuss too https://calendly.com/d/4mp-gd3-k5k/litellm-1-1-onboarding-chat

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants