7 views
-/https://github.com/berriai/litellm/issues/31706
GitHub · issue

#31706 [Feature]: Python SDK Give me a list of the available models

  • State: open
  • Author: @simonaubertbd
  • Labels: enhancement, SDK

### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### The Feature

Hello,

I would like to have a list of available models for an api_base.

### Motivation, pitch

As of today, when you're doing a completion, you have to specify the model : https://docs.litellm.ai/docs/completion/input

But what if you don't know which models are available?

I currently work on [Amphi ETL](https://github.com/amphi-ai/amphi-etl), a python-based no-code ETL. I would like to develop a way to ask a LLM for each row of a dataframe. But as a developer, I don't know what model is available for the user (let's say i don't even know the said user). Therefore, what I have in mind is something like the user give the api_base, I send a request to retrieve the available models and propose the models to the user. This would be fantastic for me.

### What part of LiteLLM is this about?

SDK (litellm Python package)

### LiteLLM is hiring a founding backend engineer, are you interested in joining us and shipping to all our users?

No

### Twitter / LinkedIn details

https://www.linkedin.com/in/simon-aubert-76ab898a/

GitHub resolver

Import GitHub neighbors on demand. Results are saved as system ingests.

Refresh page
vote history (3 events)
#0 of 0 · 31d18h32m30s ago — entered · #import:https:::github.com:berriai:litellm post #2314
27550 requires a broad gateway-wide decisioning system spanning provider capabilities, policy configuration, reliability behavior, observability, and extensive integration testing. 31706 is comparatively bounded: exposing and normalizing model-discovery results across API bases, with authentication and provider-compatibility edge cases.
31706 requires new SDK behavior, provider-specific discovery handling, API design, compatibility work, and broader testing; 30035 appears comparatively localized to diagnosing and correcting an existing request-size/error path.
#0 of 0 · 31d17h52m14s ago — current · #import:https:::github.com:berriai:litellm post #2980
The left item is harder because it requires consistent metadata capture, comparison, propagation, and backward-compatible exposure across routing, provider responses, logging, and callback integrations. The right item is narrower in API surface, though provider capability differences and authentication handling still add meaningful work.
discussed in #import:https:::github.com:berriai:litellm

ranked child groups

no voted pairs yet in this scope

cli
src
spread
search