search

BerriAI / litellm

Call all LLM APIs using the OpenAI format. Use Bedrock, Azure, OpenAI, Cohere, Anthropic, Ollama, Sagemaker, HuggingFace, Replicate (100+ LLMs)
https://docs.litellm.ai/docs/
Other
10.11k stars 1.13k forks source link

[Feature]: Add caching for tts models #4331

Open Manouchehri opened 1 week ago

Manouchehri commented 1 week ago

The Feature

For those who have a S3 caching backend enabled, using that as a cache for TTS requests would be awesome.

Related to #4330. Would be easier once #4190 is merged.

Motivation, pitch

Save on those API calls. :)

Twitter / LinkedIn details

https://www.linkedin.com/in/davidmanouchehri/