%%capture
import os, re
if "COLAB_" not in "".join(os.environ.keys()):
    !pip install unsloth
else:
    # Do this only in Colab notebooks! Otherwise use pip install unsloth
    import torch; v = re.match(r"[0-9]{1,}\.[0-9]{1,}", str(torch.__version__)).group(0)
    xformers = "xformers==" + ("0.0.33.post1" if v=="2.9" else "0.0.32.post2" if v=="2.8" else "0.0.29.post3")
    !pip install --no-deps bitsandbytes accelerate {xformers} peft trl triton cut_cross_entropy unsloth_zoo
    !pip install sentencepiece protobuf "datasets==4.3.0" "huggingface_hub>=0.34.0" hf_transfer
    !pip install --no-deps unsloth
!pip install transformers==4.56.2
!pip install --no-deps trl==0.22.2

Then, you can test with this code

Developed by: hmuegyi
License: apache-2.0
Finetuned from model: unsloth/qwen2.5-7b-bnb-4bit

python

from unsloth import FastLanguageModel
import torch

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "hmuegyi/Qwen2.5-7B-bnb-en-my-alt", 
    max_seq_length = 2048,
    load_in_4bit = True, # Memory သက်သာအောင်
)
FastLanguageModel.for_inference(model)

alpaca_prompt = """### Instruction:
You are a professional English-Burmese translator. 
Detect the input language and provide the translation in the opposite language.

### Input:
{}

### Response:
{}"""

input_text = "I love Myanmar Country."  # you can change input text
inputs = tokenizer(
    [
        alpaca_prompt.format(
            input_text, # Input
            "",         # Response
        )
    ], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs,
                         max_new_tokens = 128,
                         temperature = 0.1,  
                         top_p = 0.5,
                         use_cache = True)

response = tokenizer.batch_decode(outputs)

final_output = response[0].split("### Response:")[1].replace(tokenizer.eos_token, "").strip()
print(f"Input: {input_text}")
print(f"Translation: {final_output}")

This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.