SA

Sakura 13B v0.11Pre

SakuraGGUF by KungerUpdated 2024-05-02
Download on Hugging Face
Parameters14.5B
Context4K
Layers40
Architectureorion
Smallest quant5.5 GB
Quick answer

Sakura 13B v0.11Pre needs about 16.0 GB of memory at Q4_K (a 8.8 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 6 quants, from 5.5 GB (Q2_K) to 29.0 GB (F16).

Sakura 13B v0.11Pre is a 14.5B-parameter Sakura model in GGUF format, quantized by Kunger. Its trained context length is 4K tokens. The repository has 6 quant files, from 5.5 GB (Q2_K) to 29.0 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K5.5 GB 9.4 GB
Too big Get file →
Q4_K8.8 GB 12.7 GB
Too big Get file →
Q4_K8.8 GB 12.7 GB
Too big Get file →
Q6_K11.9 GB 15.8 GB
Too big Get file →
—29.0 GB 32.9 GB
Too big Get file →
F1629.0 GB 32.9 GB
Too big Get file →

What GPU do you need for Sakura 13B v0.11Pre?

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: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q4_K (8.8 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (11.9 GB file, about 19.1 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (11.9 GB file, about 19.1 GB in use).
  • 48 GB VRAM: the best quant that fits is F16 (29.0 GB file, about 36.2 GB in use).

Which Sakura 13B v0.11Pre quant should you choose?

Q4_K (8.8 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 Q2_K at 5.5 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureorion
Parameters14.5B
Layers40
Trained context4,096 tokens
KV cache (fp16)800 KB per token

Source

Quantized byKunger
FamilySakura
Downloads (HF)40
Last updated2024-05-02

Frequently asked questions about Sakura 13B v0.11Pre

How much VRAM does Sakura 13B v0.11Pre need?

About 16.0 GB at Q4_K with an 8K context window. That is the 8.8 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 13B v0.11Pre on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Sakura 13B v0.11Pre GGUF?

It depends on the quant: Q2_K is 5.5 GB, Q4_K is 8.8 GB, Q4_K is 8.8 GB, Q6_K is 11.9 GB, file is 29.0 GB, F16 is 29.0 GB.

What context length does Sakura 13B v0.11Pre support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 6.7 GB at 8K and 3.4 GB at 4K.

Where can I download Sakura 13B v0.11Pre GGUF?

The files are on Hugging Face in the Kunger/Sakura-13B-v0.11Pre-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 13B v0.11Pre 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 Kunger

All by Kunger →

More Sakura models

All Sakura →

Other medium models

All medium models →