NVIDIA · CC BY 4.0
Parakeet TDT 0.6B v3
NVIDIA’s fast transcription model for 25 European languages, with word timings.
How much memory it needs
Memory to run each model, from its model card or the size of its files. Smaller versions trade some quality for less memory.
| Model | Version | Needs | Business use |
|---|---|---|---|
| Parakeet-TDT 0.6B v3 | 8-bit GGUF | 0.7 GB | Allowed |
| FP32 | 2.5 GB |
Which computers run it
The best model in this group for each machine.
| Computer | Best version | Fit |
|---|---|---|
| Laptop or PC, no graphics card, 32 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| RTX 4060 · 8 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| RTX 3060 · 12 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| RTX 5060 Ti · 16 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| RTX 4090 · 24 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| RTX 5090 · 32 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| Mac with M4 · 16 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| Mac with M4 Pro · 48 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| Mac with M5 Max · 128 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| Ryzen AI Max+ 395 · 128 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| DGX Spark · 128 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| RTX PRO 6000 · 96 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| Mac Studio with M5 Ultra · 256 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
| Mac Studio with M3 Ultra · 512 GB | Parakeet-TDT 0.6B v3 FP32 | Runs great |
What it is good and bad at
Good at
- Fast
- Word timestamps
- Runs on CPU (8-bit GGUF)
Not good at
- European languages only
- CC BY 4.0: credit required
Start it
- Parakeet-TDT 0.6B v3: runs in NeMo-Speech.cpp (GGUF) (also NeMo (PyTorch), Transformers, MLX (mlx-community), CoreML (FluidInference)). Model card