MI

long mistral

MistralGGUF by JobaulaUpdated 2024-10-07
Download on Hugging Face
Parameters22.2B
Context128K
Layers56
Architecturellama
Smallest quant44.5 GB
Quick answer

long mistral needs about 46.9 GB of memory at F16 (a 44.5 GB file) with an 8K context window, so it fits on a 48 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 7.5 GB.

long mistral is a 22.2B-parameter Mistral model in GGUF format, quantized by Jobaula. Its trained context length is 128K tokens. The repository has 1 quant file, from 44.5 GB (F16) to 44.5 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
F1644.5 GB 46.9 GB
Too big Get file →

What GPU do you need for long mistral?

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: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: the best quant that fits is F16 (44.5 GB file, about 46.9 GB in use).

Which long mistral quant should you choose?

F16 (44.5 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters22.2B
Layers56
Trained context131,072 tokens
KV cache (fp16)224 KB per token

Source

Quantized byJobaula
FamilyMistral
Licenseapache-2.0
Downloads (HF)7
Last updated2024-10-07

Frequently asked questions about long mistral

How much VRAM does long mistral need?

About 46.9 GB at F16 with an 8K context window. That is the 44.5 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 long mistral on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 46.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of long mistral GGUF?

It depends on the quant: F16 is 44.5 GB.

What context length does long mistral support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.9 GB at 8K and 7.5 GB at 32K.

Where can I download long mistral GGUF?

The files are on Hugging Face in the Jobaula/long_mistral repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does long mistral 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.

Similar models

Browse all →

More from Jobaula

All by Jobaula →

More Mistral models

All Mistral →

Other medium models

All medium models →