---
license: mit
language:
- en
library_name: adapter-transformers
---

xriminact/TarsChattyBasev0.1 GGUF models converted using llama.cpp

| Name        | Quant Method           | Bits  | Max RAM Required  | Usecase  |
| ----------- |:----------------------:|:-----:|:-----------------:|:--------:|
| TarsChattyBasev0.1-Q4_K_M.gguf       | Q4_K_M | 4 | ~6.5 GB | balanced quality
| TarsChattyBasev0.1-Q5_K_M.gguf       | Q5_K_M | 5 | ~7.5 GB | large, very low quality loss
| TarsChattyBasev0.1-Q8_0.gguf         | Q8_0   | 8 | ~10.5 GB | very large, extremely low quality loss