Skip to content

Sarvamai · Transcription

Saaras V3

SarvamaiTranscriptionSarvam family
From Fastrouter's listing:
Saaras V3 is Sarvam AI's next-generation automatic speech recognition (ASR) model engineered for Indian languages and real-world speech conditions. Saaras V3 supports 23 languages (22 official Indian languages plus English) within a unified multilingual model. The model achieves approximately 19% Word Error Rate (WER) on the IndicVoices benchmark, improving significantly from its predecessor Saaras V2.5 which had ~22% WER. On the 10 most popular languages in the IndicVoices dataset, it achieves 19.3% WER, and performance advantages widen for lower-resource Indian languages
Market median price
$0.0053 /min
Max context
2K
Listings
1
Providers
1

Evidence and sources

Released 10 Feb 2026
Sources: FastRouter data, llminfo researchPrimary data from FastRouter from llminfo research: release date

Advantages and trade-offs

Compared with 10 representative transcription models
We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 4 maker's prices; 2 market median prices.

Input & output

as stated by the model's providers
Audio
Text

Listings

 Fastroutersarvam/saaras:v3Source: FastRouter data2K$0.0053

Benchmarks

No benchmark scores yet.

Also known as: sarvam/saaras:v3

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works