GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ardhikakrsh
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
ardhikakrsh
Model Tree
mistralai/Mistral-7B-Instruct-v0.3
Input Modalities
Output Modalities
Supported Functionality
base_model: mistralai/Mistral-7B-Instruct-v0.3 language:
This is a LoRA adapter fine-tuned from the Mistral-7B-Instruct-v0.3 model, specifically designed to generate English Question-Answer Pairs (QAPs) aligned with the Kurikulum Merdeka for Phase E and Phase F.
The model is trained and adapted for Indonesian senior high school (SMA) students following the Kurikulum Merdeka standards.
This model serves as the core inference engine for a mobile application designed to help high school teachers automatically generate reading comprehension exercises, including:
Since this is a LoRA adapter (PEFT), it must be loaded together with its base model. Below is an example using the `transformers` and `peft` libraries:
```python from peft import PeftModel from transformers import AutoModelForCausalLM, AutoTokenizer
base_model_id = "mistralai/Mistral-7B-Instruct-v0.3" base_model = AutoModelForCausalLM.from_pretrained( base_model_id, load_in_4bit=True ) tokenizer = AutoTokenizer.from_pretrained(base_model_id)
adapter_id = "ardhikakrsh/mistral-merdeka-curriculum-sma" model = PeftModel.from_pretrained(base_model, adapter_id)
print("Model is ready for question generation!")
Download them in the Files & versions tab.