MI

mixtral 8x22b v0.1 imatrix

MistralGGUF by ChristianAzinnUpdated 2024-04-12
Download on Hugging Face
Parameters141B
Context64K
Layers56
Architecturellama
Smallest quant37.9 GB
Quick answer

mixtral 8x22b v0.1 imatrix needs about 88.0 GB of memory at Q4_K_M (a 85.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 12 quants, from 37.9 GB (IQ2_XXS) to 115.5 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 7.5 GB.

mixtral 8x22b v0.1 imatrix is a 141B-parameter Mistral model in GGUF format, quantized by ChristianAzinn. Its trained context length is 64K tokens. The repository has 12 quant files, from 37.9 GB (IQ2_XXS) to 115.5 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_XXS37.9 GB 40.3 GB
Too big Get file →
IQ2_XS42.0 GB 44.4 GB
Too big Get file →
Q2_K52.1 GB (5 parts) 54.5 GB
Too big Get file →
IQ3_XXS54.9 GB (5 parts) 57.3 GB
Too big Get file →
Q3_K_S61.5 GB (5 parts) 63.9 GB
Too big Get file →
Q3_K_M67.8 GB (5 parts) 70.2 GB
Too big Get file →
Q3_K_L72.6 GB (5 parts) 75.0 GB
Too big Get file →
Q4_K_S80.5 GB (6 parts) 82.9 GB
Too big Get file →
Q4_K_M85.6 GB (6 parts) 88.0 GB
Too big Get file →
Q5_K_S97.0 GB (6 parts) 99.4 GB
Too big Get file →
Q5_K_M100.0 GB (6 parts) 102.3 GB
Too big Get file →
Q6_K115.5 GB (9 parts) 117.9 GB
Too big Get file →

What GPU do you need for mixtral 8x22b v0.1 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: the best quant that fits is IQ2_XS (42.0 GB file, about 44.4 GB in use).

Which mixtral 8x22b v0.1 imatrix quant should you choose?

Q4_K_M is the usual starting point for mixtral 8x22b v0.1 imatrix: it balances file size and quality well. Here it is a 85.6 GB file.

If memory is tight, the smallest file is IQ2_XXS at 37.9 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
Parameters141B (approximate, mixture of experts)
Layers56
Trained context65,536 tokens
KV cache (fp16)224 KB per token

Source

Quantized byChristianAzinn
FamilyMistral
Licenseapache-2.0
Downloads (HF)500
Last updated2024-04-12

Frequently asked questions about mixtral 8x22b v0.1 imatrix

How much VRAM does mixtral 8x22b v0.1 imatrix need?

About 88.0 GB at Q4_K_M with an 8K context window. That is the 85.6 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 mixtral 8x22b v0.1 imatrix on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 40.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of mixtral 8x22b v0.1 imatrix GGUF?

It depends on the quant: IQ2_XXS is 37.9 GB, IQ2_XS is 42.0 GB, Q2_K is 52.1 GB, IQ3_XXS is 54.9 GB, Q3_K_S is 61.5 GB, Q3_K_M is 67.8 GB and 6 more.

What context length does mixtral 8x22b v0.1 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 mixtral 8x22b v0.1 imatrix GGUF?

The files are on Hugging Face in the ChristianAzinn/mixtral-8x22b-v0.1-imatrix repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does mixtral 8x22b v0.1 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 ChristianAzinn

All by ChristianAzinn →

More Mistral models

All Mistral →

Other large models

All large models →