Vytre Core - Intelligence Engine for VytreAWOS
Vytre Core is a specialized, lightweight AI model fine-tuned for enterprise workforce operations. Optimized for organizational structure creation, agent deployment, workflow planning, and task execution—not general internet chat or creative storytelling.
Model Details
- Developer: Vytre
- Base Model: NousResearch/Hermes-3-Llama-3-8B
- Architecture: LoRA/QLoRA fine-tuned Llama-3
- License: Apache-2.0
- Use Case: Enterprise AI workforce operations
How to Use
1. Install Dependencies
pip install transformers torch peft accelerate bitsandbytes
2. Run Inference
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_name = "tarvico/vytre-core"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")
prompt = """You are Vytre, an enterprise workforce operating intelligence model.
Input: Create a marketing department
Output: """
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
Intended Uses
- ✅ Organization structure creation
- ✅ AI agent definition and deployment
- ✅ Workflow planning and generation
- ✅ Task decomposition and routing
- ✅ Governance and approval checks
- ✅ Enterprise tool execution routing
Out of Scope
- ❌ General internet chat
- ❌ Creative storytelling
- ❌ Open-ended casual conversation
Training Dataset
Synthetic enterprise operations dataset covering:
- Organization creation
- Agent creation
- Workflow generation
- Task decomposition
- Governance checks
- Tool usage
Contact
For enterprise inquiries: [your-email@example.com]