base_model: Snowflake/snowflake-arctic-base
language:
- en
library_name: transformers
license: apache-2.0
quantized_by: mradermacher
tags:
- snowflake
- arctic
- moe
About
weighted/imatrix quants of https://huggingface.co/Snowflake/snowflake-arctic-base
static quants are available at https://huggingface.co/mradermacher/snowflake-arctic-base-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 | 96.4 | for the desperate |
| PART 1 PART 2 PART 3 | i1-IQ1_M | 107.4 | mostly desperate |
| PART 1 PART 2 PART 3 | i1-IQ2_XXS | 125.7 | |
| PART 1 PART 2 PART 3 | i1-IQ2_XS | 140.4 | |
| PART 1 PART 2 PART 3 | i1-IQ2_S | 141.6 | |
| PART 1 PART 2 PART 3 PART 4 | i1-IQ2_M | 156.2 | |
| PART 1 PART 2 PART 3 PART 4 | i1-Q2_K_S | 162.6 | very low quality |
| PART 1 PART 2 PART 3 PART 4 | i1-Q2_K | 174.8 | IQ3_XXS probably better |
| PART 1 PART 2 PART 3 PART 4 | i1-IQ3_XXS | 184.5 | lower quality |
| PART 1 PART 2 PART 3 PART 4 | i1-IQ3_XS | 196.1 | |
| P1 P2 P3 P4 P5 | i1-IQ3_S | 207.3 | beats Q3_K* |
| P1 P2 P3 P4 P5 | i1-Q3_K_S | 207.3 | IQ3_XS probably better |
| P1 P2 P3 P4 P5 | i1-IQ3_M | 210.1 | |
| P1 P2 P3 P4 P5 | i1-Q3_K_M | 229.9 | IQ3_S probably better |
| P1 P2 P3 P4 P5 P6 | i1-Q3_K_L | 248.9 | IQ3_M probably better |
| P1 P2 P3 P4 P5 P6 | i1-IQ4_XS | 256.2 | |
| P1 P2 P3 P4 P5 P6 | i1-Q4_0 | 272.3 | fast, low quality |
| P1 P2 P3 P4 P5 P6 | i1-Q4_K_S | 273.5 | optimal size/speed/quality |
| P1 P2 P3 P4 P5 P6 | i1-Q4_K_M | 291.2 | fast, recommended |
| P1 P2 P3 P4 P5 P6 P7 | i1-Q4_1 | 301.3 | |
| P1 P2 P3 P4 P5 P6 P7 | i1-Q5_K_S | 331.4 | |
| P1 P2 P3 P4 P5 P6 P7 | i1-Q5_K_M | 341.7 | |
| P1 P2 P3 P4 P5 P6 P7 P8 P9 | i1-Q6_K | 395.3 | 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
FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized.
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. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.