Downloads · 30 days
21
2% of all-time downloads
RichardErkhov/LeoLM_-_leo-hessianai-70b-gguf
LeoLM_-_leo-hessianai-70b-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
21
2% of all-time downloads
All-time downloads
1.3K
Public
Repo size
778 GB
Likes
0
Public
Click a slice to open those files.
.gguf902 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
leo-hessianai-70b - GGUF
| Name | Quant method | Size |
|---|---|---|
| leo-hessianai-70b.Q2_K.gguf | Q2_K | 23.71GB |
| leo-hessianai-70b.IQ3_XS.gguf | IQ3_XS | 26.37GB |
| leo-hessianai-70b.IQ3_S.gguf | IQ3_S | 27.86GB |
| leo-hessianai-70b.Q3_K_S.gguf | Q3_K_S | 27.86GB |
| leo-hessianai-70b.IQ3_M.gguf | IQ3_M | 28.82GB |
| leo-hessianai-70b.Q3_K.gguf | Q3_K | 30.99GB |
| leo-hessianai-70b.Q3_K_M.gguf | Q3_K_M | 30.99GB |
| leo-hessianai-70b.Q3_K_L.gguf | Q3_K_L | 33.67GB |
| leo-hessianai-70b.IQ4_XS.gguf | IQ4_XS | 34.64GB |
| leo-hessianai-70b.Q4_0.gguf | Q4_0 | 36.2GB |
| leo-hessianai-70b.IQ4_NL.gguf | IQ4_NL | 36.55GB |
| leo-hessianai-70b.Q4_K_S.gguf | Q4_K_S | 36.55GB |
| leo-hessianai-70b.Q4_K.gguf | Q4_K | 38.58GB |
| leo-hessianai-70b.Q4_K_M.gguf | Q4_K_M | 38.58GB |
| leo-hessianai-70b.Q4_1.gguf | Q4_1 | 40.2GB |
| leo-hessianai-70b.Q5_0.gguf | Q5_0 | 44.2GB |
| leo-hessianai-70b.Q5_K_S.gguf | Q5_K_S | 44.2GB |
| leo-hessianai-70b.Q5_K.gguf | Q5_K | 45.41GB |
| leo-hessianai-70b.Q5_K_M.gguf | Q5_K_M | 45.41GB |
| leo-hessianai-70b.Q5_1.gguf | Q5_1 | 48.2GB |
| leo-hessianai-70b.Q6_K.gguf | Q6_K | 52.7GB |
| leo-hessianai-70b.Q8_0.gguf | Q8_0 | 68.26GB |
datasets:
Meet LeoLM, the first open and commercially available German Foundation Language Model built on Llama-2.
Our models extend Llama-2's capabilities into German through continued pretraining on a large corpus of German-language and mostly locality specific text.
Thanks to a compute grant at HessianAI's new supercomputer 42, we release a series foundation models trained with 8k context length
under the Llama-2 community license. Now, we're finally releasing the
much anticipated leo-hessianai-70b, the largest model of this series based on Llama-2-70b.
With this release, we hope to bring a new wave of opportunities to German open-source and commercial LLM research and accelerate adoption.
Read our blog post or our paper (preprint coming soon) for more details!
A project by Björn Plüster and Christoph Schuhmann in collaboration with LAION and HessianAI.
First install direct dependencies:
pip install transformers torch
Then load the model in transformers. Note that this requires lots of VRAM and most-likely multiple devices. Use load_in_8bit=True or load_in_4bit=True
to save some memory by using a quantized version. For more quantized versions, check out our models at TheBloke's page: (coming soon!)
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model = AutoModelForCausalLM.from_pretrained(
model="LeoLM/leo-hessianai-70b",
device_map="auto",
torch_dtype=torch.bfloat16,
use_flash_attention_2=False # Set to true to use FA2. Requires `pip install flash-attn`
)

