metadata
language:
- en
library_name: transformers
license: apache-2.0
quantized_by: mradermacher
tags:
- moe
About
weighted/imatrix quants of https://huggingface.co/mistralai/Mixtral-8x7B-v0.1
Usage
If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files.
Provided Quants
Link | Type | Size/GB | Notes |
---|---|---|---|
GGUF | i1-IQ1_S | 9.7 | |
GGUF | i1-IQ2_XXS | 12.5 | |
GGUF | i1-IQ2_XS | 13.8 | |
GGUF | i1-Q2_K | 17.5 | |
GGUF | i1-IQ3_XXS | 18.5 | fast, lower quality |
GGUF | i1-Q3_K_XS | 19.2 | IQ3-XXS probably better |
GGUF | i1-Q3_K_S | 20.6 | IQ3-XXS probably better |
GGUF | i1-Q3_K_M | 22.7 | lower quality |
GGUF | i1-Q3_K_L | 24.3 | |
GGUF | i1-Q4_K_S | 26.9 | fast, medium quality |
GGUF | i1-Q4_K_M | 28.6 | fast, medium quality |
GGUF | i1-Q5_K_M | 33.4 | best weighted quant |