from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Load model
model_name = "afzalur/case-study-mistral-7b-full"
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

# Generate case study
prompt = "Create a case study about sustainable business practices for an MBA course"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(
    **inputs, 
    max_new_tokens=512, 
    temperature=0.7,
    do_sample=True
)

result = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(result[len(prompt):])  # Only show generated content

🎯 Business Applications

Business Schools: Curriculum development
Corporate Training: Leadership scenarios
Consulting: Case study libraries
EdTech: Automated content generation

📊 Model Details

Base Model: Mistral 7B Instruct v0.3
Training: LoRA fine-tuning on business case studies
Context Length: 8K tokens
Model Size: ~13.5GB
Quality: Professional academic standard

💻 API Usage

python

# For programmatic access
from transformers import pipeline

generator = pipeline(
    "text-generation",
    model="afzalur/case-study-mistral-7b-full",
    torch_dtype=torch.float16,
    device_map="auto"
)

result = generator(
    "Create a case study about digital transformation",
    max_new_tokens=512,
    temperature=0.7
)
print(result[0]['generated_text'])

🔄 Alternative Testing

If the inference widget isn't available, you can:

Download and run locally (recommended for best performance)
Use Google Colab for testing without local setup

📞 Professional Services

Available for:

Custom model development
Business education AI solutions
Model deployment and optimization
Training data curation

Contact: https://www.upwork.com/freelancers/~0198c0e561dde1229f