Instructions to use Fabisch/Dia2-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Dia2
How to use Fabisch/Dia2-2B with Dia2:
from dia2 import Dia2, GenerationConfig, SamplingConfig dia = Dia2.from_repo("Fabisch/Dia2-2B", device="cuda", dtype="bfloat16") config = GenerationConfig( cfg_scale=2.0, audio=SamplingConfig(temperature=0.8, top_k=50), use_cuda_graph=True, ) result = dia.generate("[S1] Hello Dia2!", config=config, output_wav="hello.wav", verbose=True) - Notebooks
- Google Colab
- Kaggle
Download tokenizer.json from Fabisch/Dia2-2B: direct link, hf CLI and curl.
- Browser
- Download file 3.53 MB
-
https://huggingface.co/Fabisch/Dia2-2B/resolve/main/tokenizer.json
- Command line
-
hf download hf://Fabisch/Dia2-2B/tokenizer.json
-
curl -L -o tokenizer.json https://huggingface.co/Fabisch/Dia2-2B/resolve/main/tokenizer.json
3.53 MB
File too large to display, you can check the raw version instead.