base_model: shisa-ai/ablation-196-finalsft2-shisa-v2-gemma3-27b
datasets:
- shisa-ai/shisa-v2-best-of-n-athenev2-tulu70b-llama33-only-no-sysprompt
- shisa-ai/shisa-v2-roleplaying-sft
- shisa-ai/translation_set_april_6
- shisa-ai/rewild-set-deepseek-subset
- shisa-ai/magpie-ultra-set
- shisa-ai/magpie-advanced-questions-set
- shisa-ai/japan-magpie-set
- shisa-ai/shisa-v2-instruction-following-sft
language:
- en
library_name: transformers
license: gemma
quantized_by: mradermacher
tags:
- generated_from_trainer
About
weighted/imatrix quants of https://huggingface.co/shisa-ai/ablation-196-finalsft2-shisa-v2-gemma3-27b
static quants are available at https://huggingface.co/mradermacher/ablation-196-finalsft2-shisa-v2-gemma3-27b-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 | 6.4 | for the desperate |
| GGUF | i1-IQ1_M | 6.9 | mostly desperate |
| GGUF | i1-IQ2_XXS | 7.8 | |
| GGUF | i1-IQ2_XS | 8.5 | |
| GGUF | i1-IQ2_S | 8.9 | |
| GGUF | i1-IQ2_M | 9.6 | |
| GGUF | i1-Q2_K_S | 9.9 | very low quality |
| GGUF | i1-Q2_K | 10.6 | IQ3_XXS probably better |
| GGUF | i1-IQ3_XXS | 10.8 | lower quality |
| GGUF | i1-IQ3_XS | 11.7 | |
| GGUF | i1-IQ3_S | 12.3 | beats Q3_K* |
| GGUF | i1-Q3_K_S | 12.3 | IQ3_XS probably better |
| GGUF | i1-IQ3_M | 12.6 | |
| GGUF | i1-Q3_K_M | 13.5 | IQ3_S probably better |
| GGUF | i1-Q3_K_L | 14.6 | IQ3_M probably better |
| GGUF | i1-IQ4_XS | 14.9 | |
| GGUF | i1-Q4_0 | 15.7 | fast, low quality |
| GGUF | i1-Q4_K_S | 15.8 | optimal size/speed/quality |
| GGUF | i1-Q4_K_M | 16.6 | fast, recommended |
| GGUF | i1-Q4_1 | 17.3 | |
| GGUF | i1-Q5_K_S | 18.9 | |
| GGUF | i1-Q5_K_M | 19.4 | |
| GGUF | i1-Q6_K | 22.3 | 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.