GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
AdhamAshraf
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
AdhamAshraf
Model Tree
aubmindlab/aragpt2-medium
Input Modalities
Output Modalities
Supported Functionality
text
dialect: {input} ↔ msa:The model generates the Modern Standard Arabic translation autoregressively.
| Parameter | Value |
|---|---|
| Batch size | 8 (effective 32) |
| Learning rate | 5e-5 |
| Scheduler | Cosine |
| Warmup | 10% |
| Gradient clipping | 1.0 |
| Parameter | Value |
|---|---|
| Temperature | 0.7 |
| Top-k | 50 |
| Top-p | 0.92 |
| Repetition penalty | 1.3 |
| Metric | Base AraGPT-2 | SlangGPT |
|---|---|---|
| chrF | 10.62 | 29.08 |
| BLEU | 0.02 | 6.63 |
| chrF Improvement | — | +18.46 (+173%) |
bash
pip install transformers torchpython
from transformers import AutoTokenizer, AutoModelForCausalLMimport torch model_name = "AdhamAshraf/SlangGPT" tokenizer = AutoTokenizer.from_pretrained(model_name) if tokenizer.pad_token is None: tokenizer.pad_token = tokenizer.eos_token tokenizer.padding_side = "left" model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype=torch.float16, device_map="auto") model.eval()python
def translate(egyptian_text): prompt = f"dialect: {egyptian_text.strip()} ↔ msa:" inputs = tokenizer( prompt, return_tensors="pt", truncation=True, max_length=64 ) inputs = { k: v.to(model.device) for k, v in inputs.items() } with torch.no_grad(): outputs = model.generate( **inputs, max_new_tokens=64, do_sample=True, temperature=0.7, top_k=50, top_p=0.92, repetition_penalty=1.3, pad_token_id=tokenizer.pad_token_id, eos_token_id=tokenizer.eos_token_id, ) full = tokenizer.decode( outputs[0], skip_special_tokens=True ) if "msa:" in full: return full.split("msa:")[-1].strip() return fullpython
print(translate("يلا فين؟"))# هيا، أين أنت؟ print(translate("إنت رايح فين؟"))# أين أنت ذاهب؟ print(translate("عايز اكل"))# أريد الطعامTry the live demo here:
https://huggingface.co/spaces/AdhamAshraf/SlangGPT
The Space allows users to:
SlangGPT was fine-tuned using:
Dataset statistics:
| Property | Value |
|---|---|
| Total samples | 18,250 |
| Format | Parallel Egyptian ↔ MSA |
| Train split | 80% |
| Validation split | 10% |
| Test split | 10% |
The dataset was derived from the original Egyptian-English corpus by Abdalrahmankamel, with English translations replaced by curated MSA equivalents.
The model was evaluated using:
User feedback collected through the Gradio Space is publicly stored in:
https://huggingface.co/datasets/AdhamAshraf/slanggpt-feedback-dataset
This feedback dataset supports:
This project is released under the MIT License.
Free for academic and commercial use with attribution.
bibtex
@software{slanggpt2026, author = {Abdelrahman Ahmed and Adham Ashraf and Ahmed Fekry}, title = {SlangGPT: Fine-tuning AraGPT-2 for Egyptian Arabic Dialect-to-MSA Translation}, year = {2026}, url = {https://github.com/adhamashraf7788/SlangGPT}} @dataset{egyptian_2_arabic, author = {Adham Ashraf and Abdelrahman Ahmed and Ahmed Fekry}, title = {Egyptian Arabic Slang to Formal Arabic Dataset}, year = {2026}, publisher = {Hugging Face}, url = {https://huggingface.co/datasets/AdhamAshraf/egyptian-2-arabic}}For bugs, issues, or feature requests:
from transformers import AutoTokenizer, AutoModelForCausalLMimport torch model_name = "AdhamAshraf/SlangGPT" tokenizer = AutoTokenizer.from_pretrained(model_name) if tokenizer.pad_token is None: tokenizer.pad_token = tokenizer.eos_token tokenizer.padding_side = "left" model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype=torch.float16, device_map="auto") model.eval()def translate(egyptian_text): prompt = f"dialect: {egyptian_text.strip()} ↔ msa:" inputs = tokenizer( prompt, return_tensors="pt", truncation=True, max_length=64 ) inputs = { k: v.to(model.device) for k, v in inputs.items() } with torch.no_grad(): outputs = model.generate( **inputs, max_new_tokens=64, do_sample=True, temperature=0.7, top_k=50, top_p=0.92, repetition_penalty=1.3, pad_token_id=tokenizer.pad_token_id, eos_token_id=tokenizer.eos_token_id, ) full = tokenizer.decode( outputs[0], skip_special_tokens=True ) if "msa:" in full: return full.split("msa:")[-1].strip() return fullprint(translate("يلا فين؟"))# هيا، أين أنت؟ print(translate("إنت رايح فين؟"))# أين أنت ذاهب؟ print(translate("عايز اكل"))# أريد الطعام@software{slanggpt2026, author = {Abdelrahman Ahmed and Adham Ashraf and Ahmed Fekry}, title = {SlangGPT: Fine-tuning AraGPT-2 for Egyptian Arabic Dialect-to-MSA Translation}, year = {2026}, url = {https://github.com/adhamashraf7788/SlangGPT}} @dataset{egyptian_2_arabic, author = {Adham Ashraf and Abdelrahman Ahmed and Ahmed Fekry}, title = {Egyptian Arabic Slang to Formal Arabic Dataset}, year = {2026}, publisher = {Hugging Face}, url = {https://huggingface.co/datasets/AdhamAshraf/egyptian-2-arabic}}