language:
- en
library_name: transformers
license: llama2
quantized_by: mradermacher
tags:
- moe
- moerge
About
weighted quants of https://huggingface.co/ibivibiv/giant-hydra-moe-240b
static quants are available at https://huggingface.co/mradermacher/giant-hydra-moe-240b-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 |
|---|---|---|---|
| PART 1 PART 2 | i1-IQ1_S | 49.5 | for the desperate |
| PART 1 PART 2 | i1-IQ2_XXS | 63.4 | |
| PART 1 PART 2 | i1-IQ2_XS | 70.5 | |
| PART 1 PART 2 | i1-IQ2_S | 72.2 | |
| PART 1 PART 2 | i1-IQ2_M | 79.2 | |
| PART 1 PART 2 | i1-Q2_K | 87.6 | IQ3_XXS probably better |
| PART 1 PART 2 | i1-IQ3_XXS | 92.7 | fast, lower quality |
| PART 1 PART 2 | i1-Q3_K_XS | 96.5 | |
| PART 1 PART 2 PART 3 | i1-IQ3_XS | 97.4 | |
| PART 1 PART 2 PART 3 | i1-Q3_K_S | 103.5 | IQ3_XS probably better |
| PART 1 PART 2 PART 3 | i1-IQ3_S | 103.5 | fast, beats Q3_K* |
| PART 1 PART 2 PART 3 | i1-IQ3_M | 105.5 | |
| PART 1 PART 2 PART 3 | i1-Q3_K_M | 114.8 | IQ3_S probably better |
| PART 1 PART 2 PART 3 | i1-Q3_K_L | 124.3 | IQ3_M probably better |
| PART 1 PART 2 PART 3 | i1-Q4_K_S | 136.2 | optimal size/speed/quality |
| PART 1 PART 2 PART 3 | i1-Q4_K_M | 144.8 | fast, medium quality |
| PART 1 PART 2 PART 3 PART 4 | i1-Q5_K_S | 164.7 | |
| PART 1 PART 2 PART 3 PART 4 | i1-Q5_K_M | 169.7 | |
| PART 1 PART 2 PART 3 PART 4 PART 5 | i1-Q6_K | 196.3 | practically like static Q6_K |
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
Thanks
I thank my company, nethype GmbH, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.