FineTune needs about 15.9 GB of memory at Q4_K_S (a 12.1 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 8 quants, from 4.1 GB (Q5_K_S) to 40.9 GB (BF16).
FineTune is a 20.4B-parameter Finetune model in GGUF format, quantized by chis767. Its trained context length is 512 tokens. The repository has 8 quant files, from 4.1 GB (Q5_K_S) to 40.9 GB (BF16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q5_K_S | 4.1 GB | 4.8 GB | Fits | Get file → | |
| Q4_K_S | 4.8 GB | 5.5 GB | Fits | Get file → | |
| Q4_K_S | 8.7 GB | 9.4 GB | Too big | Get file → | |
| Q4_K_S | 12.1 GB | 12.8 GB | Too big | Get file → | |
| BF16 | 15.2 GB | 15.9 GB | Too big | Get file → | |
| Q8_0 | 15.4 GB | 16.1 GB | Too big | Get file → | |
| Q8_0 | 15.4 GB | 16.1 GB | Too big | Get file → | |
| BF16 | 40.9 GB | 41.6 GB | Too big | Get file → |
What GPU do you need for FineTune?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) 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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: the best quant that fits is Q5_K_S (4.1 GB file, about 7.8 GB in use).
- 12 GB VRAM: the best quant that fits is Q4_K_S (4.8 GB file, about 8.5 GB in use).
- 16 GB VRAM: the best quant that fits is Q4_K_S (12.1 GB file, about 15.9 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (15.4 GB file, about 19.1 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (15.4 GB file, about 19.1 GB in use).
- 48 GB VRAM: the best quant that fits is BF16 (40.9 GB file, about 44.6 GB in use).
Which FineTune quant should you choose?
Q4_K_S (12.1 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
If memory is tight, the smallest file is Q5_K_S at 4.1 GB, with the quality trade-offs that come with fewer bits. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about FineTune
How much VRAM does FineTune need?
About 15.9 GB at Q4_K_S with an 8K context window. That is the 12.1 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 FineTune on an 8 GB GPU?
Yes, with Q5_K_S: the 4.1 GB file needs about 7.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of FineTune GGUF?
It depends on the quant: Q5_K_S is 4.1 GB, Q4_K_S is 4.8 GB, Q4_K_S is 8.7 GB, Q4_K_S is 12.1 GB, BF16 is 15.2 GB, Q8_0 is 15.4 GB and 2 more.
What context length does FineTune support?
The GGUF header lists a trained context length of 512 tokens (512). The KV cache grows with the context you use: about 3.2 GB at 8K and 0.2 GB at 512.
Where can I download FineTune GGUF?
The files are on Hugging Face in the chis767/FineTune repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.