Speech-to-text
Parakeet TDT 0.6B v3 requirements, license & speed
By NVIDIA · released 2025-08-04 · 600M parameters
Commercial useYes
Min memoryAt least 2 GB RAM to load
Runs on CPUUnknown
StreamingYes
Key facts
- License
- cc-by-4.0 source
- Commercial use
- Yes sourceCC-BY-4.0 (attribution); card: ready for commercial/non-commercial use
- Parameters
- 600M sourceThis model has 600 million model parameters
- Minimum memory
- At least 2 GB RAM to load source
- Runs on CPU
- Unknown
- Streaming
- Yes sourceChunked streaming via NeMo script
- Voice cloning
- n/a (speech-to-text)
- Languages
- en, es, fr, de, bg, hr, cs, da, nl, et, fi, el, hu, it, lv, lt, mt, pl, pt, ro, sk, sl, sv, ru, uk source
- Released
- 2025-08-04 sourceHugging Face repo creation date (proxy for release date)
- Gated on HF
- No
Measured speed & memory
Benchmark pending. We haven't measured Parakeet TDT 0.6B v3 yet. Planned targets: Raspberry Pi 5, CPU-only x86, Apple Silicon Mac and consumer GPUs. Get notified.
Accuracy
We don't run accuracy evals. For word error rate, check the Hugging Face Open ASR Leaderboard.
Comparisons: Whisper vs Parakeet
Similar STT models
Spotted an error? Report it. Every fact links to the model card or repo it came from.