MI

mixtral8x22 imatrix

MistralGGUF by nistenUpdated 2024-04-11
Download on Hugging Face
Parameters141B
Context64K
Layers56
Architecturellama
Smallest quant46.2 GB
Quick answer

mixtral8x22 imatrix needs about 66.2 GB of memory at its main file (a 63.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 5 quants, from 46.2 GB to 149.4 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 7.5 GB.

mixtral8x22 imatrix is a 141B-parameter Mistral model in GGUF format, quantized by nisten. Its trained context length is 64K tokens. The repository has 5 quant files, from 46.2 GB to 149.4 GB.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—46.2 GB 48.5 GB
Too big Get file →
—49.9 GB 52.3 GB
Too big Get file →
—63.8 GB (5 parts) 66.2 GB
Too big Get file →
—77.2 GB (5 parts) 79.6 GB
Too big Get file →
—149.4 GB (10 parts) 151.8 GB
Too big Get file →

What GPU do you need for mixtral8x22 imatrix?

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: no quant fits at an 8K context.

Which mixtral8x22 imatrix quant should you choose?

If memory is tight, the smallest file is the first one in the table at 46.2 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters141B (approximate, mixture of experts)
Layers56
Trained context65,536 tokens
KV cache (fp16)224 KB per token

Source

Quantized bynisten
FamilyMistral
Licenseapache-2.0
Downloads (HF)83
Last updated2024-04-11

Frequently asked questions about mixtral8x22 imatrix

How much VRAM does mixtral8x22 imatrix need?

About 66.2 GB at its main file with an 8K context window. That is the 63.8 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 mixtral8x22 imatrix on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 48.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of mixtral8x22 imatrix GGUF?

It depends on the quant: file is 46.2 GB, file is 49.9 GB, file is 63.8 GB, file is 77.2 GB, file is 149.4 GB.

What context length does mixtral8x22 imatrix support?

The GGUF header lists a trained context length of 65,536 tokens (64K). 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 mixtral8x22 imatrix GGUF?

The files are on Hugging Face in the nisten/mixtral8x22-imatrix-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 mixtral8x22 imatrix 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 Mixtral-8x22B-v0.1

Similar models

Browse all →

More from nisten

All by nisten →

More Mistral models

All Mistral →

Other large models

All large models →