Skip to content

Add saaras:v4 as the default and latest Sarvam STT model - #6900

Open
dhruvladia-sarvam wants to merge 1 commit into
livekit:mainfrom
dhruvladia-sarvam:feat/add-saaras-v4-stt-model
Open

Add saaras:v4 as the default and latest Sarvam STT model#6900
dhruvladia-sarvam wants to merge 1 commit into
livekit:mainfrom
dhruvladia-sarvam:feat/add-saaras-v4-stt-model

Conversation

@dhruvladia-sarvam

Copy link
Copy Markdown
Contributor

What

  • Adds saaras:v4 to the SarvamSTTModels type literal and MODEL_CONFIGS
  • Changes the default model from saarika:v2.5 to saaras:v4 in SarvamSTTOptions and STT.init
  • Updates all docstrings, comments, and the external docs page to reflect the new model and default

Capabilities

saaras:v4 shares the same capability surface as saaras:v3:

  • Full 23-language coverage
  • Mode selection (transcribe, translate, verbatim, translit, codemix)
  • Fine-grained VAD parameters
  • Standard speech-to-text endpoint (no translate endpoint)

Behaviour change

Callers that construct sarvam.STT() without an explicit model= argument will now use saaras:v4 instead of saarika:v2.5. Callers that already pass model= are unaffected.

@dhruvladia-sarvam
dhruvladia-sarvam requested a review from a team as a code owner August 19, 2026 08:49

@devin-ai-integration devin-ai-integration Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no potential bugs to report.

View in Devin Review to see 1 additional finding.

Open in Devin Review

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant