Open ASR · AMI-Cleaned WER ↗English short-form; normalization unversioned
9.42% WERnvidia/parakeet-tdt-0.6b-v3
nvidiaListed May 27, 2026
NVIDIA: Parakeet TDT 0.6B v3
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...
Measured performance
All leaderboards ↗Benchmark results
Word error rate · lower is better
Open ASR · Earnings22-Cleaned-AA-chunked WER ↗English short-form; normalization unversioned
5.85% WERnvidia/parakeet-tdt-0.6b-v3
Open ASR · Gigaspeech-Cleaned WER ↗English short-form; normalization unversioned
7.99% WERnvidia/parakeet-tdt-0.6b-v3
Open ASR · LS Clean WER ↗English short-form; normalization unversioned
1.52% WERnvidia/parakeet-tdt-0.6b-v3
Open ASR · LS Other WER ↗English short-form; normalization unversioned
3.13% WERnvidia/parakeet-tdt-0.6b-v3
Open ASR · SPGISpeech WER ↗English short-form; normalization unversioned
3.63% WERnvidia/parakeet-tdt-0.6b-v3
Open ASR · URGENT2024 WER ↗English short-form; normalization unversioned
11.52% WERnvidia/parakeet-tdt-0.6b-v3
Open ASR · Voice Arena Monsoon WER ↗English short-form; normalization unversioned
4.14% WERnvidia/parakeet-tdt-0.6b-v3
Open ASR · Voxpopuli-AA-Cleaned WER ↗English short-form; normalization unversioned
3.05% WERnvidia/parakeet-tdt-0.6b-v3
Access & cost
1 quotePrices across providers
- Input
- $25.00
- Output
- $0.00
Model specifications
- Developer
- nvidia
- Modality
- Language
- Context
- Not reported
- Maximum output
- Not reported
- Weights
- Not reported
- Provider listed
- May 27, 2026
- First discovered
- Oct 16, 2026
Exact identities
- openrouter
nvidia/parakeet-tdt-0.6b-v3 - open-asr
nvidia%2Fparakeet-tdt-0.6b-v3
Sources
Updated Oct 20, 2026
More from nvidia
View all ↗nvidiaListed 4 days ago
NVIDIA: Nemotron 3.5 Super VL (free)
NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...
1M contextImage inputVideo inputNo reported score
No reported score
nvidiaListed 5 days ago
NVIDIA Nemotron 3.5 Super Vision
Ask NVIDIA Nemotron 3.5 Super questions about an image: describe scenes, read text and charts, and reason over what it sees.
No reported score
No reported score
Price not listed
nvidiaListed 5 days ago
NVIDIA Nemotron 3.5 Super Text
Chat and reason with NVIDIA Nemotron 3.5 Super, a hybrid Mamba-Transformer mixture-of-experts model with optional step-by-step thinking.
No reported score
No reported score
Price not listed
nvidiaListed 7 days ago
NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4
1M context120BImage inputOpen weightsNo reported score
No reported score