RichardErkhov/MadShift_-_Maral-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Maral - GGUF

• Model creator: https://huggingface.co/MadShift/
• Original model: https://huggingface.co/MadShift/Maral/

Name	Quant method	Size
Maral.Q2_K.gguf	Q2_K	2.96GB
Maral.IQ3_XS.gguf	IQ3_XS	3.28GB
Maral.IQ3_S.gguf	IQ3_S	3.43GB
Maral.Q3_K_S.gguf	Q3_K_S	3.41GB
Maral.IQ3_M.gguf	IQ3_M	3.52GB
Maral.Q3_K.gguf	Q3_K	3.74GB
Maral.Q3_K_M.gguf	Q3_K_M	3.74GB
Maral.Q3_K_L.gguf	Q3_K_L	4.03GB
Maral.IQ4_XS.gguf	IQ4_XS	4.18GB
Maral.Q4_0.gguf	Q4_0	4.34GB
Maral.IQ4_NL.gguf	IQ4_NL	4.38GB
Maral.Q4_K_S.gguf	Q4_K_S	4.37GB
Maral.Q4_K.gguf	Q4_K	4.58GB
Maral.Q4_K_M.gguf	Q4_K_M	4.58GB
Maral.Q4_1.gguf	Q4_1	4.78GB
Maral.Q5_0.gguf	Q5_0	5.21GB
Maral.Q5_K_S.gguf	Q5_K_S	5.21GB
Maral.Q5_K.gguf	Q5_K	5.34GB
Maral.Q5_K_M.gguf	Q5_K_M	5.34GB
Maral.Q5_1.gguf	Q5_1	5.65GB
Maral.Q6_K.gguf	Q6_K	6.14GB
Maral.Q8_0.gguf	Q8_0	7.95GB

Original model description:

language: - ru license: apache-2.0 pipeline_tag: text-generation library_name: transformers

Maral

Description

Maral is a general-purpose generative language model that demonstrates excellent performance in tasks such as summarization and question-answering, specifically in the Russian language. Its advanced capabilities allow it to generate coherent and contextually accurate responses, making it highly effective for a wide range of natural language processing applications.

👨‍💻 Examples of usage

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

tokenizer = AutoTokenizer.from_pretrained("MadShift/Maral")
model = AutoModelForCausalLM.from_pretrained("MadShift/Maral", device_map="auto")

input_text = "Введите свой текст здесь"
input_ids = tokenizer(input_text, return_tensors="pt").to(model.device)

outputs = model.generate(**input_ids, max_new_tokens=256)
print(tokenizer.decode(outputs[0]))