SA

Sakura SOLAR Instruct

SakuraGGUF by andrijdavidUpdated 2023-12-28
Download on Hugging Face
Parameters10.7B
Context4K
Layers48
Architecturellama
Smallest quant4.5 GB
Quick answer

Sakura SOLAR Instruct needs about 8.6 GB of memory at Q4_K_M (a 6.5 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 18 quants, from 4.5 GB (Q8_0) to 21.5 GB (F16).

Sakura SOLAR Instruct is a 10.7B-parameter Sakura model in GGUF format, quantized by andrijdavid. Its trained context length is 4K tokens. The repository has 18 quant files, from 4.5 GB (Q8_0) to 21.5 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_04.5 GB 5.8 GB
Fits Get file →
Q2_K4.5 GB 5.9 GB
Fits Get file →
Q3_K_S4.7 GB 6.0 GB
Fits Get file →
Q3_K5.2 GB 6.5 GB
Fits Get file →
Q3_K_M5.2 GB 6.5 GB
Fits Get file →
Q3_K_L5.7 GB 7.0 GB
Fits Get file →
Q4_06.1 GB 7.4 GB
Tight Get file →
Q4_K_S6.1 GB 7.4 GB
Tight Get file →
Q4_K6.5 GB 7.8 GB
Tight Get file →
Q4_K_M6.5 GB 7.8 GB
Tight Get file →
Q4_16.7 GB 8.0 GB
Too big Get file →
Q5_07.4 GB 8.7 GB
Too big Get file →
Q5_K_S7.4 GB 8.7 GB
Too big Get file →
Q5_K7.6 GB 8.9 GB
Too big Get file →
Q5_K_M7.6 GB 8.9 GB
Too big Get file →
Q5_18.1 GB 9.4 GB
Too big Get file →
Q6_K8.8 GB 10.1 GB
Too big Get file →
F1621.5 GB 22.8 GB
Too big Get file →

What GPU do you need for Sakura SOLAR Instruct?

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 Q3_K_L (5.7 GB file, about 7.8 GB in use).
  • 12 GB VRAM: the best quant that fits is Q6_K (8.8 GB file, about 10.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q6_K (8.8 GB file, about 10.9 GB in use).
  • 24 GB VRAM: the best quant that fits is F16 (21.5 GB file, about 23.6 GB in use).

Which Sakura SOLAR Instruct quant should you choose?

Q4_K_M is the usual starting point for Sakura SOLAR Instruct: it balances file size and quality well. Here it is a 6.5 GB file.

The Q8_0 file is 4.5 GB, about 0.7 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.

If memory is tight, the smallest file is Q8_0 at 4.5 GB, with the quality trade-offs that come with fewer bits. 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

Architecturellama
Parameters10.7B
Layers48
Trained context4,096 tokens
KV cache (fp16)192 KB per token

Source

Quantized byandrijdavid
FamilySakura
Downloads (HF)141
Last updated2023-12-28

Frequently asked questions about Sakura SOLAR Instruct

How much VRAM does Sakura SOLAR Instruct need?

About 8.6 GB at Q4_K_M with an 8K context window. That is the 6.5 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 Sakura SOLAR Instruct on an 8 GB GPU?

Yes, with Q3_K_L: the 5.7 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 Sakura SOLAR Instruct GGUF?

It depends on the quant: Q8_0 is 4.5 GB, Q2_K is 4.5 GB, Q3_K_S is 4.7 GB, Q3_K is 5.2 GB, Q3_K_M is 5.2 GB, Q3_K_L is 5.7 GB and 12 more.

What context length does Sakura SOLAR Instruct support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 1.6 GB at 8K and 0.8 GB at 4K.

Where can I download Sakura SOLAR Instruct GGUF?

The files are on Hugging Face in the andrijdavid/Sakura-SOLAR-Instruct-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 Sakura SOLAR Instruct use?

Hugging Face lists the license as cc-by-nc-sa-4.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Similar models

Browse all →

More from andrijdavid

All by andrijdavid →

More Sakura models

All Sakura →

Other medium models

All medium models →