metadata
language:
- en
library_name: transformers
license: llama2
quantized_by: mradermacher
tags:
- moe
About
static quants of https://huggingface.co/ibivibiv/orthorus-125b-moe
weighted/imatrix quants available at https://huggingface.co/mradermacher/orthorus-125b-moe-i1-GGUF
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
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|---|---|---|---|
| GGUF | Q2_K | 46.7 | |
| PART 1 PART 2 | Q3_K_XS | 51.5 | |
| PART 1 PART 2 | Q3_K_S | 55.0 | |
| PART 1 PART 2 | Q3_K_M | 61.0 | lower quality |
| PART 1 PART 2 | Q3_K_L | 66.0 | |
| PART 1 PART 2 | Q4_K_S | 72.1 | fast, medium quality |
| PART 1 PART 2 | Q4_K_M | 76.4 | fast, medium quality |
| PART 1 PART 2 | Q5_K_S | 87.1 | |
| PART 1 PART 2 | Q5_K_M | 89.6 | |
| PART 1 PART 2 PART 3 | Q6_K | 103.7 | very good quality |
| PART 1 PART 2 PART 3 | Q8_0 | 134.0 | fast, best quality |
Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):
And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9