datasets:
- tiiuae/falcon-refinedweb
- pankajmathur/WizardLM_Orca
language:
- en
library_name: transformers
quantized_by: mradermacher
About
weighted/imatrix quants of https://huggingface.co/quantumaikr/falcon-180B-WizardLM_Orca
static quants are available at https://huggingface.co/mradermacher/falcon-180B-WizardLM_Orca-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 | i1-IQ1_S | 37.4 | for the desperate |
| GGUF | i1-IQ2_XXS | 46.8 | |
| PART 1 PART 2 | i1-IQ2_XS | 52.1 | |
| PART 1 PART 2 | i1-IQ2_S | 55.6 | |
| PART 1 PART 2 | i1-IQ2_M | 60.3 | |
| PART 1 PART 2 | i1-Q2_K | 65.9 | IQ3_XXS probably better |
| PART 1 PART 2 | i1-IQ3_XXS | 68.5 | fast, lower quality |
| PART 1 PART 2 | i1-IQ3_XS | 74.4 | |
| PART 1 PART 2 | i1-IQ3_S | 76.8 | fast, beats Q3_K* |
| PART 1 PART 2 | i1-Q3_K_S | 76.8 | IQ3_XS probably better |
| PART 1 PART 2 | i1-IQ3_M | 80.5 | |
| PART 1 PART 2 | i1-Q3_K_M | 84.6 | IQ3_S probably better |
| PART 1 PART 2 | i1-Q3_K_L | 91.1 | IQ3_M probably better |
| PART 1 PART 2 PART 3 | i1-Q4_K_S | 100.6 | optimal size/speed/quality |
| PART 1 PART 2 PART 3 | i1-Q4_K_M | 107.9 | fast, medium quality |
| PART 1 PART 2 PART 3 | i1-Q5_K_S | 122.9 | |
| PART 1 PART 2 PART 3 | i1-Q5_K_M | 130.1 | |
| PART 1 PART 2 PART 3 PART 4 | i1-Q6_K | 146.6 | 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.