36 lines
1.1 KiB
Markdown
36 lines
1.1 KiB
Markdown
---
|
|
language:
|
|
- multilingual
|
|
- pl
|
|
- en
|
|
license: apache-2.0
|
|
library_name: gguf
|
|
tags:
|
|
- gguf
|
|
- quantization
|
|
inference: false
|
|
pipeline_tag: text-generation
|
|
base_model: speakleash/Bielik-PL-11B-v3.0-Instruct
|
|
base_model_relation: quantized
|
|
---
|
|
|
|
# Bielik-PL-11B-v3.0-Instruct-GGUF
|
|
|
|
This repository contains community-quantized GGUF files for the [Bielik-PL-11B-v3.0-Instruct](https://huggingface.co/speakleash/Bielik-PL-11B-v3.0-Instruct) model by SpeakLeash.
|
|
|
|
The distinctive feature of this "-PL-" version is the optimized tokenizer (APT4) specifically dedicated to the Polish language.
|
|
|
|
## Available formats:
|
|
* **`Bielik-PL-11B-v3.0-Instruct.f16.gguf`**
|
|
* **`Bielik-PL-11B-v3.0-Instruct.bf16.gguf`**
|
|
* **`Bielik-PL-11B-v3.0-Instruct.Q8_0.gguf`**
|
|
* **`Bielik-PL-11B-v3.0-Instruct.Q6_K.gguf`**
|
|
* **`Bielik-PL-11B-v3.0-Instruct.Q5_K_M.gguf`**
|
|
* **`Bielik-PL-11B-v3.0-Instruct.Q4_K_M.gguf`**
|
|
|
|
---
|
|
|
|
## Quantization Author
|
|
Files quantized by **macszym** using [llama.cpp](https://github.com/ggerganov/llama.cpp).
|
|
*Thanks to [SpeakLeash](https://speakleash.org/) for the original model!*
|