FishAudio S2 needs about 1.6 GB of memory at Q4_K_M (a 0.9 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 13 quants, from 0.9 GB (Q4_K_M) to 9.9 GB (F16).
FishAudio S2 is a Fishaudio model in GGUF format, quantized by niobures. The repository has 13 quant files, from 0.9 GB (Q4_K_M) to 9.9 GB (F16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.9 GB | 1.6 GB | Fits | Get file → | |
| Q8_0 | 1.0 GB | 1.6 GB | Fits | Get file → | |
| F16 | 1.4 GB | 2.1 GB | Fits | Get file → | |
| Q2_K | 2.6 GB | 3.4 GB | Fits | Get file → | |
| Q4_K_M | 2.8 GB | 3.6 GB | Fits | Get file → | |
| Q3_K | 3.0 GB | 3.9 GB | Fits | Get file → | |
| Q4_K_M | 3.6 GB | 4.5 GB | Fits | Get file → | |
| Q5_K_M | 4.0 GB | 5.0 GB | Fits | Get file → | |
| Q6_K | 4.5 GB | 5.6 GB | Fits | Get file → | |
| Q8_0 | 5.3 GB | 6.5 GB | Fits | Get file → | |
| Q8_0 | 5.6 GB | 6.8 GB | Fits | Get file → | |
| F16 | 9.1 GB | 10.7 GB | Too big | Get file → | |
| F16 | 9.9 GB | 11.6 GB | Too big | Get file → |
What GPU do you need for FishAudio S2?
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 Q3_K (3.0 GB file, about 3.9 GB in use).
- 6 GB VRAM: the best quant that fits is Q6_K (4.5 GB file, about 5.6 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (5.6 GB file, about 6.8 GB in use).
- 12 GB VRAM: the best quant that fits is F16 (9.9 GB file, about 11.6 GB in use).
Which FishAudio S2 quant should you choose?
Q4_K_M is the usual starting point for FishAudio S2: it balances file size and quality well. Here it is a 0.9 GB file.
The Q8_0 file is 1.0 GB, about 1.1 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about FishAudio S2
How much VRAM does FishAudio S2 need?
About 1.6 GB at Q4_K_M with an 8K context window. That is the 0.9 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 FishAudio S2 on an 8 GB GPU?
Yes, with Q8_0: the 5.6 GB file needs about 6.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of FishAudio S2 GGUF?
It depends on the quant: Q4_K_M is 0.9 GB, Q8_0 is 1.0 GB, F16 is 1.4 GB, Q2_K is 2.6 GB, Q4_K_M is 2.8 GB, Q3_K is 3.0 GB and 7 more.
Where can I download FishAudio S2 GGUF?
The files are on Hugging Face in the niobures/FishAudio-S2 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.