Parakeet models

8 Parakeet models from NVIDIA on Hugging Face, from 609M to 1.1B parameters, published by NVIDIA in F32. The smallest official model, parakeet-ctc-0.6b, needs about 2.7 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.

Pick a size

One row per official Parakeet size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.

ModelParametersNative VRAMFP8 VRAMINT4 VRAMLive GPU fit (native)Est. $/hrKV cache at 32K
parakeet-tdt-0.6b-v3627M2.8 GB0.7 GB0.4 GBRTX 3070$0.088/hrnot published for this architecture
parakeet-ctc-1.1b1.1B4.8 GB1.2 GB0.6 GBRTX 3070$0.088/hrnot published for this architecture

VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.

Official models (4)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
parakeet-ctc-0.6b609MF322.7 GBRTX 3070$0.088/hr
parakeet-rnnt-0.6b617MF322.8 GBRTX 3070$0.088/hr
parakeet-tdt-0.6b-v3627MF322.8 GBRTX 3070$0.088/hr
parakeet-ctc-1.1b1.1BF324.8 GBRTX 3070$0.088/hr

Fine-tunes and community models (4)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
parakeet-tdt-0.6b-v2618MF322.8 GBRTX 3070$0.088/hr
parakeet-tdt-0.6b-v3627MF322.8 GBRTX 3070$0.088/hr
fluid-speech-pro627MF322.8 GBRTX 3070$0.088/hr
parakeet-ultra627MF161.4 GBRTX 3070$0.088/hr

VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.

More from NVIDIA

Start ComfyUI on a rented GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.