kokoro de hui base needs about 0.7 GB of memory at Q8_0 (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.1 GB (Q8_0) to 0.2 GB (F16).
Kokoro de hui base – lightweight German multispeaker TTS
Kokoro de hui base is a 0.0817B parameter TTS model trained on German audio. It is a Stage‑1 multispeaker base that can be paired with German voice packs for production quality. The model is Apache‑licensed and available in GGUF format.
Good at
- German multispeaker TTS
- Fast inference on low‑memory GPUs
- Open‑weight deployment
Limits
- No vision support
- Requires 8k context for memory estimates
- No built‑in voice, needs external voice pack
Best for
- Deploying German TTS in edge devices
- Generating multilingual audio with voice packs
- Rapid prototyping of TTS services
Who it’s for. Developers building German TTS applications who need a small, open‑weight model. Researchers testing multispeaker conditioning on German data. Companies deploying TTS in production environments with Apache licensing.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q8_0 | 0.1 GB | 0.7 GB | Fits | Get file → | |
| F16 | 0.2 GB | 0.7 GB | Fits | Get file → |
For 8 GB VRAM, use Q8_0 (0.14 GB file, ~0.7 GB memory at 8k context). For slightly higher memory budgets, F16 (0.16 GB file, ~0.7 GB memory) offers similar performance. Both are safe quantizations for this backbone.
What GPU do you need for kokoro de hui base?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is F16 (0.2 GB file, about 0.7 GB in use).
Which kokoro de hui base quant should you choose?
Q8_0 (0.1 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about kokoro de hui base
How much VRAM does kokoro de hui base need?
About 0.7 GB at Q8_0 with an 8K context window. That is the 0.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run kokoro de hui base on an 8 GB GPU?
Yes, with F16: the 0.2 GB file needs about 0.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of kokoro de hui base GGUF?
It depends on the quant: Q8_0 is 0.1 GB, F16 is 0.2 GB.
Where can I download kokoro de hui base GGUF?
The files are on Hugging Face in the cstr/kokoro-de-hui-base-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does kokoro de hui base use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Kokoro de hui base on 8 GB VRAM?
Yes, the Q8_0 quantization requires about 0.7 GB of VRAM at 8k context, well within an 8 GB limit.
Does this model support voice packs?
It is a base model; it needs an external German voice pack such as df_victoria or dm_martin for production quality.
What license applies to this model?
The model is Apache‑2.0 licensed, allowing free use in commercial and personal projects.