Instructions to use markldn/Dia2-1B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Dia2
How to use markldn/Dia2-1B-GGUF with Dia2:
from dia2 import Dia2, GenerationConfig, SamplingConfig dia = Dia2.from_repo("markldn/Dia2-1B-GGUF", 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
Commit History
-
Publish F16, Q8_0 and Q4_0 variants with GPU1 verification and speed benchmarks
Verified signature22efd17 -
Document native voice cloning, PCM streaming and CPU validation
Verified signature36971cf -
Record measured CPU latency: 61.2 seconds for 10 seconds of audio
Verified signatureef463f4 -
Publish CPU-validated Dia2-1B F16 GGUF and Mimi with native audio.cpp source link
Verified signature9054fcb -
initial commit
Verified signature143576d