litellm - 💡(How to fix) Fix [Feature]: Support bedrock provider in litellm.transcription() for Mistral Voxtral [1 participants]

Check for existing issues

• I have searched the existing issues and checked that my issue is not a duplicate.

The Feature

Currently, litellm.transcription() doesn't support the bedrock provider prefix, making it impossible to use Mistral Voxtral hosted on Bedrock for audio transcription.

Requested config (example):

- model_name: SPEECH_TO_TEXT
  litellm_params:
    model: bedrock/mistral.voxtral-small-24b-2507
  model_info:
    mode: audio_transcription

The model is available on Bedrock: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-mistral-ai-voxtral-small-24b-2507.html

Mistral Voxtral transcription was added for the direct Mistral provider in #22418 — a Bedrock equivalent would follow the same pattern, and would enable AWS-native deployments to use Voxtral with IAM auth, VPC routing, and AWS cost consolidation.

Motivation, pitch

Motivation is to have best support for provider Bedrock.

What part of LiteLLM is this about?

Proxy

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

No

Twitter / LinkedIn details

No response