AuK Base and Flash needs about 3.9 GB of memory at F16 (a 3.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 9 quants, from 0.6 GB (F32) to 7.4 GB (BF16).
AuK Base and Flash – 1.5B audio model for TTS and editing
AuK Base and Flash is a 1.53B parameter foundation model for speech generation and editing. It supports zero‑shot TTS, instruction‑based TTS, content editing, acoustic editing, and paralinguistic editing through a natural‑language interface. The model is available in multiple GGUF quantizations for audio.cpp.
Good at
- Zero‑shot TTS
- Instruction‑based TTS
- Content and acoustic editing
Limits
- No vision support
- Requires 8k context for memory estimates
- MIT license limits commercial use to MIT terms
Best for
- Text‑to‑speech generation
- Speech content rewriting
- Paralinguistic voice editing
Who it’s for. Developers building speech applications that need a single model for generation and editing. Researchers testing audio generation tasks with open‑source weights. Audio engineers creating voice‑editing tools.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| F32 | 0.6 GB | 1.2 GB | Fits | Get file → | |
| Q8_0 | 1.6 GB | 2.3 GB | Fits | Get file → | |
| Q8_0 | 1.6 GB | 2.3 GB | Fits | Get file → | |
| F16 | 3.1 GB | 3.9 GB | Fits | Get file → | |
| F16 | 3.1 GB | 3.9 GB | Fits | Get file → | |
| Q8_0 | 4.3 GB | 5.3 GB | Fits | Get file → | |
| F32 | 6.1 GB | 7.4 GB | Tight | Get file → | |
| F32 | 6.1 GB | 7.4 GB | Tight | Get file → | |
| BF16 | 7.4 GB | 8.8 GB | Too big | Get file → |
For 8 GB VRAM, use Q8_0 (1.64 GB) or Q8_0 (4.25 GB) for higher quality. For 4 GB VRAM, use Q8_0 (1.64 GB). For 2 GB VRAM, use F32 (0.64 GB). Memory estimates are approximate.
What GPU do you need for AuK Base and Flash?
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 (3.1 GB file, about 3.9 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (4.3 GB file, about 5.3 GB in use).
- 8 GB VRAM: the best quant that fits is F32 (6.1 GB file, about 7.4 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (7.4 GB file, about 8.8 GB in use).
Which AuK Base and Flash quant should you choose?
F16 (3.1 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
If memory is tight, the smallest file is F32 at 0.6 GB, with the quality trade-offs that come with fewer bits. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about AuK Base and Flash
How much VRAM does AuK Base and Flash need?
About 3.9 GB at F16 with an 8K context window. That is the 3.1 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 AuK Base and Flash on an 8 GB GPU?
Yes, with F32: the 6.1 GB file needs about 7.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of AuK Base and Flash GGUF?
It depends on the quant: F32 is 0.6 GB, Q8_0 is 1.6 GB, Q8_0 is 1.6 GB, F16 is 3.1 GB, F16 is 3.1 GB, Q8_0 is 4.3 GB and 3 more.
Where can I download AuK Base and Flash GGUF?
The files are on Hugging Face in the audio-cpp/AuK-Base-and-Flash-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 AuK Base and Flash use?
Hugging Face lists the license as mit. 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 AuK Base on 4 GB VRAM?
Yes, the Q8_0 quantization at 1.64 GB fits within 4 GB VRAM, leaving room for runtime overhead.
Does AuK support image inputs?
No, AuK is an audio model and does not handle vision tasks.
What license does AuK use?
AuK is released under the MIT license, allowing free use with attribution.