Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


NorLlama-3B - GGUF
- Model creator: https://huggingface.co/NorGLM/
- Original model: https://huggingface.co/NorGLM/NorLlama-3B/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [NorLlama-3B.Q2_K.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q2_K.gguf) | Q2_K | 2.51GB |
| [NorLlama-3B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.IQ3_XS.gguf) | IQ3_XS | 2.51GB |
| [NorLlama-3B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.IQ3_S.gguf) | IQ3_S | 2.51GB |
| [NorLlama-3B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q3_K_S.gguf) | Q3_K_S | 2.51GB |
| [NorLlama-3B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.IQ3_M.gguf) | IQ3_M | 2.56GB |
| [NorLlama-3B.Q3_K.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q3_K.gguf) | Q3_K | 2.56GB |
| [NorLlama-3B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q3_K_M.gguf) | Q3_K_M | 2.56GB |
| [NorLlama-3B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q3_K_L.gguf) | Q3_K_L | 2.59GB |
| [NorLlama-3B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.IQ4_XS.gguf) | IQ4_XS | 2.51GB |
| [NorLlama-3B.Q4_0.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q4_0.gguf) | Q4_0 | 0.2GB |
| [NorLlama-3B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.IQ4_NL.gguf) | IQ4_NL | 0.49GB |
| [NorLlama-3B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q4_K_S.gguf) | Q4_K_S | 2.78GB |
| [NorLlama-3B.Q4_K.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q4_K.gguf) | Q4_K | 2.82GB |
| [NorLlama-3B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q4_K_M.gguf) | Q4_K_M | 2.82GB |
| [NorLlama-3B.Q4_1.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q4_1.gguf) | Q4_1 | 0.21GB |
| [NorLlama-3B.Q5_0.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q5_0.gguf) | Q5_0 | 0.22GB |
| [NorLlama-3B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q5_K_S.gguf) | Q5_K_S | 2.91GB |
| [NorLlama-3B.Q5_K.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q5_K.gguf) | Q5_K | 2.94GB |
| [NorLlama-3B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q5_K_M.gguf) | Q5_K_M | 2.94GB |
| [NorLlama-3B.Q5_1.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q5_1.gguf) | Q5_1 | 0.23GB |
| [NorLlama-3B.Q6_K.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q6_K.gguf) | Q6_K | 3.58GB |
| [NorLlama-3B.Q8_0.gguf](https://huggingface.co/RichardErkhov/NorGLM_-_NorLlama-3B-gguf/blob/main/NorLlama-3B.Q8_0.gguf) | Q8_0 | 0.27GB |




Original model description:
---
license: cc-by-nc-sa-4.0
language:
- 'no'
---

Gnerative Pretrained Tranformer with 3 Billion parameters for Norwegian. NorLlama-3B is based on Llama architechture, and pretrained on [Tencent Pre-training Framework](https://github.com/Tencent/TencentPretrain)

It belongs to NorGLM, a suite of pretrained Norwegian Generative Language Models. NorGLM can be used for non-commercial purposes.

## Datasets
All models in NorGLM are trained on 200G datasets, nearly 25B tokens, including Norwegian, Denish, Swedish, Germany and English.

## Run the Model

```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "NorGLM/NorLlama-3B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map='auto',
    torch_dtype=torch.bfloat16
)

text = "Tom ønsket å gå på barene med venner"
inputs = tokenizer(text, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=20)
```

## Citation Information
If you feel our work is helpful, please cite our paper:

```
@article{liu2023nlebench+,
  title={NLEBench+ NorGLM: A Comprehensive Empirical Analysis and Benchmark Dataset for Generative Language Models in Norwegian},
  author={Liu, Peng and Zhang, Lemei and Farup, Terje Nissen and Lauvrak, Even W and Ingvaldsen, Jon Espen and Eide, Simen and Gulla, Jon Atle and Yang, Zhirong},
  journal={arXiv preprint arXiv:2312.01314},
  year={2023}
}
```