GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Etherll
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Etherll
Model Tree
ibm-granite/granite-4.0-h-350m
Input Modalities
Output Modalities
Supported Functionality
Arabic Diacritization Model | نَمُوذَجُ تَشْكِيلِ النُّصُوصِ الْعَرَبِيَّةِ
نموذج بحجم 350 مليون بارامتر مخصص لتشكيل النصوص العربية. تم تدريب هذا النموذج بضبط نموذج
ibm-granite/granite-4.0-h-350m
على مجموعة البيانات
Misraj/Sadeed_Tashkeela.
python
from transformers import AutoModelForCausalLM, AutoTokenizer #تحميل النموذجmodel_id = "Etherll/Tashkeel-350M-v2"model = AutoModelForCausalLM.from_pretrained( model_id, device_map="auto", torch_dtype="bfloat16",)tokenizer = AutoTokenizer.from_pretrained(model_id) # إضافة التشكيلprompt = "السلام عليكم" input_ids = tokenizer.apply_chat_template( [{"role": "user", "content": "قم بتشكيل هذا النص "+ ":\n"+prompt}], add_generation_prompt=True, return_tensors="pt", tokenize=True,).to(model.device) output = model.generate( input_ids, do_sample=False, ) print(tokenizer.decode(output[0, input_ids.shape[-1]:], skip_special_tokens=True))السلام عليكماَلسَلَامُ عَلَيْكُمْA 350M parameter model for Arabic diacritization (Tashkeel). This model is a fine-tune of ibm-granite/granite-4.0-h-350m on the Misraj/Sadeed_Tashkeela dataset.
The Python code for usage is the same as listed in the Arabic section above.
السلام عليكماَلسَلَامُ عَلَيْكُمْThis lfm2 model was trained 2x faster with Unsloth and Huggingface's TRL library.
from transformers import AutoModelForCausalLM, AutoTokenizer #تحميل النموذجmodel_id = "Etherll/Tashkeel-350M-v2"model = AutoModelForCausalLM.from_pretrained( model_id, device_map="auto", torch_dtype="bfloat16",)tokenizer = AutoTokenizer.from_pretrained(model_id) # إضافة التشكيلprompt = "السلام عليكم" input_ids = tokenizer.apply_chat_template( [{"role": "user", "content": "قم بتشكيل هذا النص "+ ":\n"+prompt}], add_generation_prompt=True, return_tensors="pt", tokenize=True,).to(model.device) output = model.generate( input_ids, do_sample=False, ) print(tokenizer.decode(output[0, input_ids.shape[-1]:], skip_special_tokens=True))