// NVIDIA Parakeet / Canary speech · Speech-to-text
parakeet-tdt_ctc-110m: VRAM & local deployment
110M parameters. About 0.6 GB at 4-bit and 0.7 GB at BF16.
released 2024-09-17licence: Commercial OKmemory figures are estimates
01Specs
| Publisher | NVIDIA · nvidia |
|---|---|
| Family / series | NVIDIA Parakeet / Canary speech · parakeet-tdt_ctc |
| Type | Speech-to-text (automatic-speech-recognition) |
| Parameters | 110M Source: official model name |
| Licence | CC-BY-4.0 · Commercial OKPermissive licence (Apache-2.0 / MIT / CC-BY / OpenMDW): commercial use allowed. |
| Released | 2024-09-17 (Hugging Face repo created) |
02How much VRAM it needs (estimate)
Speech models are small; the weight size is close to the total memory needed.
| Precision | Weights | Weights + overhead | Smallest GPU | Mac |
|---|---|---|---|---|
| BF16 / FP16 (original) | 0.2 GB | 0.7 GB | 8GB GPU | Mac 16GB |
| FP8 / INT8 / Q8_0 | 0.1 GB | 0.6 GB | 8GB GPU | Mac 16GB |
| 4-bit (Q4_K_M / AWQ / GPTQ-Int4) | 0.1 GB | 0.6 GB | 8GB GPU | Mac 16GB |
Bits per weight: BF16 16, FP8 8.5, 4-bit 4.85. “Fits” = total ≤ 90% of memory; Mac GPU can use ~2/3 of unified memory (≤36GB) or ~3/4 (larger). For MoE, all experts must be in memory, so size by total parameters. Method →
03How to run it
Only runtimes the publisher or the runtime’s official library names for this model.
- NeMo — mentioned in the official model card
04Official repositories & quantised builds
1 official repo(s) from nvidia.
05Other NVIDIA Parakeet / Canary speech models
All 10 →| Model | Params | 4-bit total | Released |
|---|---|---|---|
| parakeet-tdt_ctc-1.1b · same series | 1.1B | 1.2 GB | 2024-05-07 |
| parakeet-unified-en-0.6b | 600M | 0.9 GB | 2026-04-07 |
| multitalker-parakeet-streaming-0.6b-v1 | 600M | 0.9 GB | 2025-10-15 |
| canary-1b-v2 | 979M | 1.2 GB | 2025-08-04 |
| canary-qwen-2.5b | 2.56B | 2.2 GB | 2025-06-26 |
| parakeet-tdt-0.6b-v2 | 600M | 0.9 GB | 2025-04-15 |
| canary-180m-flash | 180M | 0.6 GB | 2025-03-11 |
| canary-1b-flash | 811M | 1.0 GB | 2025-03-07 |
| parakeet-tdt_ctc-0.6b-ja | 600M | 0.9 GB | 2024-05-13 |
06Sources
Data checked 2025-02-18. Parameters, context and licence are copied from the official repository; memory figures are our estimates.