---
base_model: anilbhatt1/phi2-oasst-guanaco-bf16-custom
inference: false
license: mit
model_creator: anilbhatt1
model_name: phi2-oasst-guanaco-bf16-custom
pipeline_tag: text-generation
quantized_by: afrideva
tags:
- gguf
- ggml
- quantized
- q2_k
- q3_k_m
- q4_k_m
- q5_k_m
- q6_k
- q8_0
---
# anilbhatt1/phi2-oasst-guanaco-bf16-custom-GGUF

Quantized GGUF model files for [phi2-oasst-guanaco-bf16-custom](https://huggingface.co/anilbhatt1/phi2-oasst-guanaco-bf16-custom) from [anilbhatt1](https://huggingface.co/anilbhatt1)


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [phi2-oasst-guanaco-bf16-custom.fp16.gguf](https://huggingface.co/afrideva/phi2-oasst-guanaco-bf16-custom-GGUF/resolve/main/phi2-oasst-guanaco-bf16-custom.fp16.gguf) | fp16 | 5.56 GB  |
| [phi2-oasst-guanaco-bf16-custom.q2_k.gguf](https://huggingface.co/afrideva/phi2-oasst-guanaco-bf16-custom-GGUF/resolve/main/phi2-oasst-guanaco-bf16-custom.q2_k.gguf) | q2_k | 1.17 GB  |
| [phi2-oasst-guanaco-bf16-custom.q3_k_m.gguf](https://huggingface.co/afrideva/phi2-oasst-guanaco-bf16-custom-GGUF/resolve/main/phi2-oasst-guanaco-bf16-custom.q3_k_m.gguf) | q3_k_m | 1.48 GB  |
| [phi2-oasst-guanaco-bf16-custom.q4_k_m.gguf](https://huggingface.co/afrideva/phi2-oasst-guanaco-bf16-custom-GGUF/resolve/main/phi2-oasst-guanaco-bf16-custom.q4_k_m.gguf) | q4_k_m | 1.79 GB  |
| [phi2-oasst-guanaco-bf16-custom.q5_k_m.gguf](https://huggingface.co/afrideva/phi2-oasst-guanaco-bf16-custom-GGUF/resolve/main/phi2-oasst-guanaco-bf16-custom.q5_k_m.gguf) | q5_k_m | 2.07 GB  |
| [phi2-oasst-guanaco-bf16-custom.q6_k.gguf](https://huggingface.co/afrideva/phi2-oasst-guanaco-bf16-custom-GGUF/resolve/main/phi2-oasst-guanaco-bf16-custom.q6_k.gguf) | q6_k | 2.29 GB  |
| [phi2-oasst-guanaco-bf16-custom.q8_0.gguf](https://huggingface.co/afrideva/phi2-oasst-guanaco-bf16-custom-GGUF/resolve/main/phi2-oasst-guanaco-bf16-custom.q8_0.gguf) | q8_0 | 2.96 GB  |



## Original Model Card:
# Finetuned microsoft-phi2 model 

- microsoft-phi2 model finetuned on "timdettmers/openassistant-guanaco" dataset with qlora technique
- Will run on a colab T4 gpu