How to run Dia 1.6B locally

Nari Labs · Text → speech · 1.6B params · Apache-2.0 (commercial OK)

An expressive dialogue TTS that generates ultra-realistic multi-speaker conversation with non-verbal cues like laughter in one pass.

What it costs to run — $0 markup

On your own machine — $0. Runs free if you have about 10 GB of VRAM (~10GB fp16; less with offload), via Dia / Transformers.
Rent a GPU — from $0.26/hr. Fits RTX 3090 24GB at the direct vendor price ($0 markup) — pay only for the minutes you generate.
Download the weights — free. Open weights at nari-labs/Dia-1.6B.

Note: generative-media models are billed per image / per second / per minute on hosted services — not per token. Running locally or on your own rented GPU is usually far cheaper and keeps your data on your machine.

Key facts

Does	Text → speech
VRAM to run	~10 GB (~10GB fp16; less with offload)
Download	~6.4 GB
Parameters	1.6B
License	Apache-2.0 (commercial use OK)
Run with	Dia / Transformers

Get Dia 1.6B on Hugging Face →

More voice & audio models

Whisper large-v3 — Speech → text, ~10 GB VRAM
Whisper large-v3-turbo — Speech → text, ~6.0 GB VRAM
Kokoro-82M — Text → speech, ~1.0 GB VRAM
XTTS-v2 — Voice cloning, ~6.0 GB VRAM
MusicGen Large — Text → music, ~16 GB VRAM
Chatterbox — Text → speech, ~8.0 GB VRAM

Browse: all media models · chat / LLM models