GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
AdarshT
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
AdarshT
Model Tree
unsloth/Llama-3.2-3B-Instruct
Input Modalities
Output Modalities
Supported Functionality
This is a fine-tuned LoRA adapter based on the unsloth/Llama-3.2-3B-Instruct base model. It has been specifically trained on the nf-analyst/indian_recipe dataset to generate authentic Indian culinary instructions, complete with cooking times, targeted diet tags, structured ingredients, and sequential preparation steps.
python
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigfrom peft import PeftModelimport torch base_model_id = "unsloth/Llama-3.2-3B-Instruct"adapter_id = "AdarshT/Llama-3.2-3B-Indian-Recipes" # Load base model in 4-bitquantization_config = BitsAndBytesConfig(load_in_4bit=True)base_model = AutoModelForCausalLM.from_pretrained(base_model_id, quantization_config=quantization_config, device_map="auto") # Merge with your recipe adaptermodel = PeftModel.from_pretrained(base_model, adapter_id)tokenizer = AutoTokenizer.from_pretrained(adapter_id) # Prompt the modelprompt = "Recipe: Butter Chicken (Murgh Makhani)\nIngredients:\n"inputs = tokenizer(prompt, return_tensors="pt").to("cuda")outputs = model.generate(**inputs, max_new_tokens=300)print(tokenizer.decode(outputs, skip_special_tokens=True))