GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
enstazao
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
enstazao
Model Tree
unsloth/Meta-Llama-3.1-8B
Input Modalities
Output Modalities
Supported Functionality
Qalb establishes a new standard for Urdu LLMs, achieving an Overall Score of 90.34. It significantly outperforms the base model and the previous state-of-the-art.
| Task | Qalb (Ours) | Alif-1.0-Instruct | LLaMA-3.1-8B-Instruct |
|---|---|---|---|
| Overall Score | 90.34 | 87.1 | 45.7 |
| Translation | 94.41 | 89.3 | 58.9 |
| Classification | 96.38 | 93.9 | 61.4 |
| Sentiment Analysis | 95.79 | 94.3 | 54.3 |
| Ethics | 90.83 | 85.7 | 27.3 |
| Reasoning | 88.59 | 83.5 | 45.6 |
| QA (Question Answering) | 80.40 | 73.8 | 30.5 |
| Generation | 85.97 | 90.2 | 42.8 |
> Note: Scores are on a 0-100 scale. Qalb outperforms the previous best model (Alif) in 6 out of 7 categories.
The easiest way to run Qalb is using the Unsloth library, which provides 2x faster inference.
python
from unsloth import FastLanguageModelimport torch model, tokenizer = FastLanguageModel.from_pretrained( model_name = "enstazao/Qalb-1.0-8B-Instruct", max_seq_length = 2048, dtype = None, load_in_4bit = True, # <--- Currently set to use 4-bit quantization)FastLanguageModel.for_inference(model) urdu_system_prompt = "آپ ایک مددگار اور بے ضرر مصنوعی ذہانت کے اسسٹنٹ ہیں۔ آپ اردو میں سوالات کے درست جوابات دیتے ہیں۔" questions = [ "پاکستان کا قومی کھیل کیا ہے؟", "لاہور شہر کیوں مشہور ہے؟ مختصر وضاحت کریں۔", "سوال: لیاقت علی خان کون تھے؟", "کراچی کو روشنیوں کا شہر کیوں کہا جاتا ہے؟", "انگریزی میں ترجمہ کریں: 'محنت کامیابی کی کنجی ہے۔'"] print("🚀 Starting Batch Generation...\n") for user_input in questions: print(f"🔹 Question: {user_input}") # Manually Format Prompt (Llama-3 Style) prompt = f"""<|begin_of_text|><|start_header_id|>system<|end_header_id|> {urdu_system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|> {user_input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>""" inputs = tokenizer([prompt], return_tensors = "pt").to("cuda") outputs = model.generate( **inputs, max_new_tokens = 256, temperature = 0.1, top_p = 0.9, repetition_penalty = 1.1, do_sample = True, eos_token_id = [tokenizer.eos_token_id, tokenizer.convert_tokens_to_ids("<|eot_id|>")] ) response = tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True) print(f"✅ Answer: {response}") print("-" * 50)Compatible with standard transformers if Unsloth is not available.
python
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigimport torch model_name = "enstazao/Qalb-1.0-8B-Instruct"urdu_system_prompt = "آپ ایک مددگار اور بے ضرر مصنوعی ذہانت کے اسسٹنٹ ہیں۔ آپ اردو میں سوالات کے درست جوابات دیتے ہیں۔" bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_double_quant=True,) print("⏳ Loading model in 4-bit...")tokenizer = AutoTokenizer.from_pretrained(model_name)model = AutoModelForCausalLM.from_pretrained( model_name, quantization_config=bnb_config, # <--- Apply 4-bit here device_map="auto" # <--- Required for quantization) terminators = [ tokenizer.eos_token_id, tokenizer.convert_tokens_to_ids("<|eot_id|>")] questions = [ "پاکستان کا قومی کھیل کیا ہے؟", "لاہور شہر کیوں مشہور ہے؟ مختصر وضاحت کریں۔", "سوال: لیاقت علی خان کون تھے؟", "سوال: اسلام آباد شہر کے بارے میں بتائیں۔", "انگریزی میں ترجمہ کریں: 'محنت کامیابی کی کنجی ہے۔'"] print("Model Loaded. Starting Generation...\n") # 5. Loop through questionsfor user_input in questions: print(f"🔹 Question: {user_input}") prompt = f"""<|begin_of_text|><|start_header_id|>system<|end_header_id|> {urdu_system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|> {user_input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>""" input_ids = tokenizer([prompt], return_tensors="pt").to("cuda") outputs = model.generate( **input_ids, max_new_tokens = 256, temperature = 0.1, top_p = 0.9, repetition_penalty = 1.1, do_sample = True, eos_token_id = terminators ) response = tokenizer.decode(outputs[0][input_ids['input_ids'].shape[1]:], skip_special_tokens=True) print(f"✅ Answer: {response}") print("-" * 50)While Qalb has been trained to be helpful and harmless, it may still reflect biases present in the training data. Users should fact-check critical information, especially in medical, legal, or religious contexts.
If you use QALB in your research, please cite:
text
@article{qalb2025, title={Qalb: Largest State-of-the-Art Urdu Large Language Model for 230M Speakers with Systematic Continued Pre-training}, author={Hassan, Muhammad Taimoor and Ahmed, Jawad and Awais, Muhammad}, journal={arXiv preprint arXiv:2601.08141}, year={2026}, eprint={2601.08141}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={[https://arxiv.org/abs/2601.08141](https://arxiv.org/abs/2601.08141)}, doi={10.48550/arXiv.2601.08141}}from unsloth import FastLanguageModelimport torch model, tokenizer = FastLanguageModel.from_pretrained( model_name = "enstazao/Qalb-1.0-8B-Instruct", max_seq_length = 2048, dtype = None, load_in_4bit = True, # <--- Currently set to use 4-bit quantization)FastLanguageModel.for_inference(model) urdu_system_prompt = "آپ ایک مددگار اور بے ضرر مصنوعی ذہانت کے اسسٹنٹ ہیں۔ آپ اردو میں سوالات کے درست جوابات دیتے ہیں۔" questions = [ "پاکستان کا قومی کھیل کیا ہے؟", "لاہور شہر کیوں مشہور ہے؟ مختصر وضاحت کریں۔", "سوال: لیاقت علی خان کون تھے؟", "کراچی کو روشنیوں کا شہر کیوں کہا جاتا ہے؟", "انگریزی میں ترجمہ کریں: 'محنت کامیابی کی کنجی ہے۔'"] print("🚀 Starting Batch Generation...\n") for user_input in questions: print(f"🔹 Question: {user_input}") # Manually Format Prompt (Llama-3 Style) prompt = f"""<|begin_of_text|><|start_header_id|>system<|end_header_id|> {urdu_system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|> {user_input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>""" inputs = tokenizer([prompt], return_tensors = "pt").to("cuda") outputs = model.generate( **inputs, max_new_tokens = 256, temperature = 0.1, top_p = 0.9, repetition_penalty = 1.1, do_sample = True, eos_token_id = [tokenizer.eos_token_id, tokenizer.convert_tokens_to_ids("<|eot_id|>")] ) response = tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True) print(f"✅ Answer: {response}") print("-" * 50)from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigimport torch model_name = "enstazao/Qalb-1.0-8B-Instruct"urdu_system_prompt = "آپ ایک مددگار اور بے ضرر مصنوعی ذہانت کے اسسٹنٹ ہیں۔ آپ اردو میں سوالات کے درست جوابات دیتے ہیں۔" bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_double_quant=True,) print("⏳ Loading model in 4-bit...")tokenizer = AutoTokenizer.from_pretrained(model_name)model = AutoModelForCausalLM.from_pretrained( model_name, quantization_config=bnb_config, # <--- Apply 4-bit here device_map="auto" # <--- Required for quantization) terminators = [ tokenizer.eos_token_id, tokenizer.convert_tokens_to_ids("<|eot_id|>")] questions = [ "پاکستان کا قومی کھیل کیا ہے؟", "لاہور شہر کیوں مشہور ہے؟ مختصر وضاحت کریں۔", "سوال: لیاقت علی خان کون تھے؟", "سوال: اسلام آباد شہر کے بارے میں بتائیں۔", "انگریزی میں ترجمہ کریں: 'محنت کامیابی کی کنجی ہے۔'"] print("Model Loaded. Starting Generation...\n") # 5. Loop through questionsfor user_input in questions: print(f"🔹 Question: {user_input}") prompt = f"""<|begin_of_text|><|start_header_id|>system<|end_header_id|> {urdu_system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|> {user_input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>""" input_ids = tokenizer([prompt], return_tensors="pt").to("cuda") outputs = model.generate( **input_ids, max_new_tokens = 256, temperature = 0.1, top_p = 0.9, repetition_penalty = 1.1, do_sample = True, eos_token_id = terminators ) response = tokenizer.decode(outputs[0][input_ids['input_ids'].shape[1]:], skip_special_tokens=True) print(f"✅ Answer: {response}") print("-" * 50)@article{qalb2025, title={Qalb: Largest State-of-the-Art Urdu Large Language Model for 230M Speakers with Systematic Continued Pre-training}, author={Hassan, Muhammad Taimoor and Ahmed, Jawad and Awais, Muhammad}, journal={arXiv preprint arXiv:2601.08141}, year={2026}, eprint={2601.08141}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={[https://arxiv.org/abs/2601.08141](https://arxiv.org/abs/2601.08141)}, doi={10.48550/arXiv.2601.08141}}