GI

Gistral-16B

GistralGGUF by maria-aiUpdated 2024-06-06
Download on Hugging Face
Parameters16.8B
Context32K
Layers76
Architecturellama
Smallest quant9.6 GB
Quick answer

Gistral-16B needs about 12.6 GB of memory at Q4_K_S (a 9.6 GB file) with an 8K context window, so it fits on a 16 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 10.2 GB.

Gistral-16B is a 16.8B-parameter Gistral model in GGUF format, quantized by maria-ai. Its trained context length is 32K tokens. The repository has 1 quant file, from 9.6 GB (Q4_K_S) to 9.6 GB (Q4_K_S).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_S9.6 GB 12.6 GB
Too big Get file →

What GPU do you need for Gistral-16B?

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_S (9.6 GB file, about 12.6 GB in use).

Which Gistral-16B quant should you choose?

Q4_K_S (9.6 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters16.8B
Layers76
Trained context32,768 tokens
KV cache (fp16)304 KB per token

Source

Quantized bymaria-ai
FamilyGistral
Licenseapache-2.0
Downloads (HF)16
Last updated2024-06-06

Frequently asked questions about Gistral-16B

How much VRAM does Gistral-16B need?

About 12.6 GB at Q4_K_S with an 8K context window. That is the 9.6 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 Gistral-16B 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 Gistral-16B GGUF?

It depends on the quant: Q4_K_S is 9.6 GB.

What context length does Gistral-16B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.6 GB at 8K and 10.2 GB at 32K.

Where can I download Gistral-16B GGUF?

The files are on Hugging Face in the maria-ai/Gistral-16B-Q4_K_S-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 Gistral-16B use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Other GGUFs of Gistral-16B

Similar models

Browse all →

More Gistral models

All Gistral →

Other medium models

All medium models →