Originally created by @arthurGrigo on GitHub (May 5, 2024).
Describe the solution you'd like
The Open WebUI's load balancing feature, which supports multiple Ollama instances (or other OpenAI-conformant APIs), could be very useful for distributing API calls in an LLM application.
It would be helpful if Open WebUI could offer an OpenAI-conformant inference API that utilizes this load balancing.
Describe alternatives you've considered
An alternative could be to place LiteLLM between Open WebUI and Ollama but this complicates the setup for most people.
Originally created by @arthurGrigo on GitHub (May 5, 2024).
**Describe the solution you'd like**
The Open WebUI's load balancing feature, which supports multiple Ollama instances (or other OpenAI-conformant APIs), could be very useful for distributing API calls in an LLM application.
It would be helpful if Open WebUI could offer an OpenAI-conformant inference API that utilizes this load balancing.
**Describe alternatives you've considered**
An alternative could be to place LiteLLM between Open WebUI and Ollama but this complicates the setup for most people.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @arthurGrigo on GitHub (May 5, 2024).
Describe the solution you'd like
The Open WebUI's load balancing feature, which supports multiple Ollama instances (or other OpenAI-conformant APIs), could be very useful for distributing API calls in an LLM application.
It would be helpful if Open WebUI could offer an OpenAI-conformant inference API that utilizes this load balancing.
Describe alternatives you've considered
An alternative could be to place LiteLLM between Open WebUI and Ollama but this complicates the setup for most people.