Speech to text · NVIDIA

Parakeet TDT 0.6B v3

Hugging Face: nvidia/parakeet-tdt-0.6b-v3 · checked 2026-10-13

Multilingual (25 European languages) successor to v2.

Datasheet

Maker	NVIDIA
Parameters	627 MVerifiedsafetensors.total from the HF API Source
Licence	CC-BY-4.0Verified
Access	Public download
Task (HF pipeline)	automatic-speech-recognition
First published on HF	2025-08-04
Last updated on HF	2026-08-05
Weights in memory	1.3 GB at 16-bit · 0.4 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build.

Benchmark

6.51%

Open ASR Leaderboard, average WER (English short-form) · #3 of 7 scored in this category

avg 6.5143, RTFx 6227.73. Source

See the full speech to text ranking

The description summarises the model card at https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.