SA

saiga2 13b

SaigaGGUF by IlyaGusevUpdated 2023-09-26
Download on Hugging Face
Parameters13B
Context2K
Layers40
Architecturellama
Smallest quant5.4 GB
Quick answer

saiga2 13b needs about 15.1 GB of memory at Q4_K (a 7.9 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 5 quants, from 5.4 GB (Q2_K) to 13.8 GB (Q8_0).

saiga2 13b is a 13B-parameter Saiga model in GGUF format, quantized by IlyaGusev. Its trained context length is 2K tokens. The repository has 5 quant files, from 5.4 GB (Q2_K) to 13.8 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K5.4 GB 7.6 GB
Tight Get file →
Q3_K6.3 GB 8.5 GB
Too big Get file →
Q4_K7.9 GB 10.0 GB
Too big Get file →
Q5_K9.2 GB 11.4 GB
Too big Get file →
Q8_013.8 GB 16.0 GB
Too big Get file →

What GPU do you need for saiga2 13b?

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 (7.9 GB file, about 15.1 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (13.8 GB file, about 21.0 GB in use).

Which saiga2 13b quant should you choose?

Q4_K (7.9 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 5.4 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

Architecturellama
Parameters13B
Layers40
Trained context2,048 tokens
KV cache (fp16)800 KB per token

Source

Quantized byIlyaGusev
FamilySaiga
Licensellama2
Downloads (HF)310
Last updated2023-09-26

Frequently asked questions about saiga2 13b

How much VRAM does saiga2 13b need?

About 15.1 GB at Q4_K with an 8K context window. That is the 7.9 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 saiga2 13b on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of saiga2 13b GGUF?

It depends on the quant: Q2_K is 5.4 GB, Q3_K is 6.3 GB, Q4_K is 7.9 GB, Q5_K is 9.2 GB, Q8_0 is 13.8 GB.

What context length does saiga2 13b support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 6.7 GB at 8K and 1.7 GB at 2K.

Where can I download saiga2 13b GGUF?

The files are on Hugging Face in the IlyaGusev/saiga2_13b_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 saiga2 13b use?

Hugging Face lists the license as llama2. 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 IlyaGusev

All by IlyaGusev →

More Saiga models

All Saiga →

Other medium models

All medium models →