Parakeet TDT 0.6B v3 is NVIDIA's 600M-parameter multilingual speech-to-text model built on the FastConformer-TDT architecture. Trained on the Granary dataset (670,000+ hours of audio), it supports automatic language detection across all official EU languages and achieves a 6.34% average word error rate on the HuggingFace Open ASR Leaderboard. Returns transcribed text with punctuation and segment timestamps.
$0.0005
Share of requests served without a server error. Client errors (4xx) do not count.
Works with the OpenAI SDK pointed at our base URL. Every parameter is in the API reference.
Length of audio transcribed per day on NagaAI, counted since November 2026.
Parakeet TDT 0.6B v3 is NVIDIA's 600M-parameter multilingual speech-to-text model built on the FastConformer-TDT architecture.
On NagaAI, Parakeet TDT 0.6B v3 costs about $0.0005 per minute. You pay only for what you use, with no subscription.
It takes audio as input and returns transcription.
Create an API key, copy the model ID from the top of this page and send requests with the OpenAI SDK pointed at NagaAI. The API section on this page has a ready example.
Nemotron 3.5 TranscribeNvidia~$0.001 per minuteNemotron 3.5 ASR Streaming Multilingual 0.6BNvidia~$0.0001 per minuteCanary Speech-to-TextNvidia~$0.024 per minute
Parakeet UltraMoondream~$0.0005 per minute
Voxtral Small Transcribe 3MistralAI~$0.002 per minuteVoxtral Mini Transcribe 3MistralAI~$0.0018 per minuteNot enough activity data to display a chart