cstr / vibevoice-realtime-0.5b-GGUF

huggingface.co
Total runs: 3.6K
24-hour runs: 38
7-day runs: -260
30-day runs: -572
Model's Last Updated: August 02 2026
text-to-speech

Introduction of vibevoice-realtime-0.5b-GGUF

Model Details of vibevoice-realtime-0.5b-GGUF

VibeVoice-Realtime-0.5B GGUF

GGUF conversion of microsoft/VibeVoice-Realtime-0.5B for use with CrispASR .

Model variants
File Quant Size Notes
vibevoice-realtime-0.5b-tts-f16.gguf F16 2.0 GB Full precision, reference quality
vibevoice-realtime-0.5b-q8_0.gguf Q8_0 1.1 GB Near-lossless quantization
vibevoice-realtime-0.5b-q4_k.gguf Q4_K 607 MB Smallest, still perfect ASR round-trip
Voice prompts

A voice prompt is required for TTS. It contains pre-computed KV caches that establish speaker identity.

File Speaker Size
vibevoice-voice-emma.gguf Emma (female, English) 2.7 MB

Convert additional voices from the VibeVoice demo voices :

python models/convert-vibevoice-voice-to-gguf.py \
    --input en-Carter_man.pt --output vibevoice-voice-carter.gguf
Usage
# Basic TTS
crispasr --tts "Hello, how are you today?" \
    -m vibevoice-realtime-0.5b-q4_k.gguf \
    --voice vibevoice-voice-emma.gguf \
    --tts-output hello.wav

# Output: 24 kHz mono WAV
Architecture

VibeVoice-Realtime-0.5B is a streaming text-to-speech model with:

  • Base LM : 4-layer Qwen2 (text encoding with voice context)
  • TTS LM : 20-layer Qwen2 (speech conditioning, autoregressive)
  • Prediction head : 4 AdaLN + SwiGLU layers (flow matching denoiser)
  • DPM-Solver++ : 20-step 2nd-order midpoint solver (cosine schedule, v-prediction)
  • Classifier-Free Guidance : dual KV cache, cfg_scale=3.0
  • sigma-VAE decoder : 7-stage transposed ConvNeXt (3200x upsample to 24kHz)
  • EOS classifier : automatic length detection
Quality verification

All quantizations produce exact ASR round-trip matches:

Input text Parakeet ASR output
"Hello world" "Hello world."
"Hello, how are you today?" "Hello, how are you today?"
"The quick brown fox jumps over the lazy dog" "The quick brown fox jumps over the lazy dog."
"Good morning everyone" "Good morning, everyone."
Conversion
# From HuggingFace model
python models/convert-vibevoice-to-gguf.py \
    --input microsoft/VibeVoice-Realtime-0.5B \
    --output vibevoice-realtime-0.5b-tts-f16.gguf \
    --include-decoder

# Quantize
crispasr-quantize vibevoice-realtime-0.5b-tts-f16.gguf \
    vibevoice-realtime-0.5b-q4_k.gguf q4_k
License

MIT (same as the original model).

Runs of cstr vibevoice-realtime-0.5b-GGUF on huggingface.co

3.6K
Total runs
38
24-hour runs
15
3-day runs
-260
7-day runs
-572
30-day runs

More Information About vibevoice-realtime-0.5b-GGUF huggingface.co Model

More vibevoice-realtime-0.5b-GGUF license Visit here:

https://choosealicense.com/licenses/mit

vibevoice-realtime-0.5b-GGUF huggingface.co

vibevoice-realtime-0.5b-GGUF huggingface.co is an AI model on huggingface.co that provides vibevoice-realtime-0.5b-GGUF's model effect (), which can be used instantly with this cstr vibevoice-realtime-0.5b-GGUF model. huggingface.co supports a free trial of the vibevoice-realtime-0.5b-GGUF model, and also provides paid use of the vibevoice-realtime-0.5b-GGUF. Support call vibevoice-realtime-0.5b-GGUF model through api, including Node.js, Python, http.

vibevoice-realtime-0.5b-GGUF huggingface.co Url

https://huggingface.co/cstr/vibevoice-realtime-0.5b-GGUF

cstr vibevoice-realtime-0.5b-GGUF online free

vibevoice-realtime-0.5b-GGUF huggingface.co is an online trial and call api platform, which integrates vibevoice-realtime-0.5b-GGUF's modeling effects, including api services, and provides a free online trial of vibevoice-realtime-0.5b-GGUF, you can try vibevoice-realtime-0.5b-GGUF online for free by clicking the link below.

cstr vibevoice-realtime-0.5b-GGUF online free url in huggingface.co:

https://huggingface.co/cstr/vibevoice-realtime-0.5b-GGUF

vibevoice-realtime-0.5b-GGUF install

vibevoice-realtime-0.5b-GGUF is an open source model from GitHub that offers a free installation service, and any user can find vibevoice-realtime-0.5b-GGUF on GitHub to install. At the same time, huggingface.co provides the effect of vibevoice-realtime-0.5b-GGUF install, users can directly use vibevoice-realtime-0.5b-GGUF installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

vibevoice-realtime-0.5b-GGUF install url in huggingface.co:

https://huggingface.co/cstr/vibevoice-realtime-0.5b-GGUF

Url of vibevoice-realtime-0.5b-GGUF

vibevoice-realtime-0.5b-GGUF huggingface.co Url

Provider of vibevoice-realtime-0.5b-GGUF huggingface.co

cstr
ORGANIZATIONS

Other API from cstr

huggingface.co

Total runs: 5.4K
Run Growth: 2.0K
Growth Rate: 36.89%
Updated:August 14 2026
huggingface.co

Total runs: 3.1K
Run Growth: 1.4K
Growth Rate: 44.81%
Updated:August 02 2026
huggingface.co

Total runs: 2.8K
Run Growth: 467
Growth Rate: 16.78%
Updated:August 02 2026
huggingface.co

Total runs: 2.6K
Run Growth: 686
Growth Rate: 26.65%
Updated:August 02 2026
huggingface.co

Total runs: 2.4K
Run Growth: 1.2K
Growth Rate: 49.02%
Updated:August 31 2026
huggingface.co

Total runs: 2.3K
Run Growth: 103
Growth Rate: 4.45%
Updated:August 02 2026
huggingface.co

Total runs: 2.0K
Run Growth: 512
Growth Rate: 24.91%
Updated:August 02 2026
huggingface.co

Total runs: 1.7K
Run Growth: 270
Growth Rate: 15.63%
Updated:August 03 2026
huggingface.co

Total runs: 1.4K
Run Growth: -2.6K
Growth Rate: -192.27%
Updated:August 02 2026
huggingface.co

Total runs: 1.4K
Run Growth: 177
Growth Rate: 12.39%
Updated:July 21 2026
huggingface.co

Total runs: 1.1K
Run Growth: 250
Growth Rate: 22.87%
Updated:July 21 2026
huggingface.co

Total runs: 1.1K
Run Growth: 464
Growth Rate: 44.96%
Updated:August 02 2026
huggingface.co

Total runs: 1.1K
Run Growth: -1.8K
Growth Rate: -168.48%
Updated:August 02 2026
huggingface.co

Total runs: 1.0K
Run Growth: 369
Growth Rate: 36.68%
Updated:August 02 2026
huggingface.co

Total runs: 905
Run Growth: 138
Growth Rate: 15.16%
Updated:August 02 2026
huggingface.co

Total runs: 871
Run Growth: 250
Growth Rate: 28.74%
Updated:August 02 2026