ModelForgeLab
Sign in Sign up

Qwen2.5-14B Russian Instruct

Official adapter · maintained by ModelForgeLab

text-generation #russian#translation#multilingual#qwen

Improves Russian instruction following and output naturalness on Qwen2.5-14B. Trained on 12k aligned RU/EN pairs.

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-14B-Instruct")
base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-14B-Instruct")
model = PeftModel.from_pretrained(base, "modelforgelab/qwen25-14b-ru-instruct")

inputs = tokenizer("Your prompt here", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Compatibility

  • peft>=0.10
  • transformers>=4.44
  • vLLM

You might also like

Qwen2.5 14B (GGUF Q4)
text-generation public

Qwen2.5 14B (GGUF Q4)

Qwen/Qwen2.5-14B-Instruct