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.

๐Ÿ‘จโ€๐Ÿ’ป 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 = "ะ’ะฒะตะดะธั‚ะต ัะฒะพะน ั‚ะตะบัั‚ ะทะดะตััŒ"
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]))
Downloads last month
6
Safetensors
Model size
8.03B params
Tensor type
BF16
ยท
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for MadShift/Maral

Quantizations
1 model