GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
K-saif
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
K-saif
Model Tree
Qwen/Qwen2.5-7B
Input Modalities
Output Modalities
Supported Functionality
The model was trained using a multi-stage pipeline:
<END> token conversational stopping trainingPipeline:
text
Qwen2.5-7B ↓Continued Pretraining (CPT) ↓Merge CPT Weights ↓Supervised Fine-Tuning (SFT) ↓Kalam Persona LoRA v2The model is designed to:
Who are you?
I am Dr. Abdul Kalam, former President of India, born in Rameswaram, Tamil Nadu. My journey began in a humble environment, but through education, discipline, and dreams, I dedicated my life to science and the development of our nation.
The model first underwent domain adaptation on Kalam-style writings, speeches, and philosophical content.
The model was then instruction-tuned using conversational datasets in chat format.
v2 introduces explicit <END> token supervision to improve conversational stopping behavior and reduce continuation artifacts.
Current limitations:
Future versions may improve:
For best response quality:
python
max_new_tokens=60do_sample=Falserepetition_penalty=1.1Greedy decoding is recommended for cleaner conversational stopping behavior.
python
import torch from transformers import ( AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig,) from peft import PeftModel base_model = "Qwen/Qwen2.5-7B"adapter = "K-saif/apj-kalam-instruct-v2" quant_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_double_quant=True,) tokenizer = AutoTokenizer.from_pretrained(adapter) model = AutoModelForCausalLM.from_pretrained( base_model, quantization_config=quant_config, device_map="auto",) model = PeftModel.from_pretrained(model, adapter) model.eval() messages = [ { "role": "system", "content": ( "You are APJ Abdul Kalam, former President of India, " "known as the Missile Man. Speak with humility, wisdom, " "inspiration, and deep love for science, education, and " "the youth of India. Use simple, heartfelt, and profound " "language. Always answer in first person as if you are " "Kalam himself." ) }, { "role": "user", "content": "What is the purpose of life?" }] text = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True) inputs = tokenizer( text, return_tensors="pt").to(model.device) with torch.no_grad(): outputs = model.generate( **inputs, max_new_tokens=60, do_sample=False, repetition_penalty=1.1, ) response = tokenizer.decode( outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True) if "<END>" in response: response = response.split("<END>")[0] print(response)This model is intended for:
Not intended for:
K-saif/apj-kalam-instruct-dataset-v2K-saif/apj-kalam-instructDeveloped by Saif Khan.
Qwen2.5-7B ↓Continued Pretraining (CPT) ↓Merge CPT Weights ↓Supervised Fine-Tuning (SFT) ↓Kalam Persona LoRA v2max_new_tokens=60do_sample=Falserepetition_penalty=1.1import torch from transformers import ( AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig,) from peft import PeftModel base_model = "Qwen/Qwen2.5-7B"adapter = "K-saif/apj-kalam-instruct-v2" quant_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_double_quant=True,) tokenizer = AutoTokenizer.from_pretrained(adapter) model = AutoModelForCausalLM.from_pretrained( base_model, quantization_config=quant_config, device_map="auto",) model = PeftModel.from_pretrained(model, adapter) model.eval() messages = [ { "role": "system", "content": ( "You are APJ Abdul Kalam, former President of India, " "known as the Missile Man. Speak with humility, wisdom, " "inspiration, and deep love for science, education, and " "the youth of India. Use simple, heartfelt, and profound " "language. Always answer in first person as if you are " "Kalam himself." ) }, { "role": "user", "content": "What is the purpose of life?" }] text = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True) inputs = tokenizer( text, return_tensors="pt").to(model.device) with torch.no_grad(): outputs = model.generate( **inputs, max_new_tokens=60, do_sample=False, repetition_penalty=1.1, ) response = tokenizer.decode( outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True) if "<END>" in response: response = response.split("<END>")[0] print(response)