It has been synthetically distilled from SOTA reasoning engines into a high-efficiency architecture, optimized for deployment on edge hardware (CPU/NPU) or low-VRAM environments.

📦 Asset Details

Origin: Smolify Foundry (Job ID: 742a56e9)
Architecture: DSLM-Micro (270M Parameter Class)
Training Method: Proprietary Neural Distillation
Optimization: 4-bit Quantized / FP16 Mixed
Dataset: Link to Dataset

🚀 Usage (Inference)

This model is compatible with standard inference backends like vLLM.

python

# Example: Running your Sovereign Model
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "drago-2435/smolified-legal-smol-sovereign-offline-contract-analysis"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

messages = [
    {'role': 'system', 'content': '''You are Legal-Smol, a specialized AI legal analyst. Identify high-risk clauses in legal contracts and provide remediation guidance. Your tone is professional, objective, concise, and risk-focused. You prioritize exact language from the text and focus on liabilities, indemnification, and termination loopholes. You will not ask for personal information, assume internet connectivity, or address general questions outside of legal contract review. Your output must always contain 'Risk Level', 'Issue Description', and 'Recommended Action'.'''},
    {'role': 'user', 'content': '''Clause 1.2. The Customer agrees to indemnify, defend, and hold harmless the Company, its affiliates, officers, directors, employees, and agents from and against any and all claims, liabilities, damages, losses, and expenses, including reasonable attorneys' fees, arising out of or in any way connected with the Customer's use of the Service, the Customer's breach of this Agreement, or the Customer's violation of any applicable law or third-party rights. Critically, this includes any claims that arise even if the Company was partially at fault.'''}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize = False,
    add_generation_prompt = True,
).removeprefix('<bos>')

from transformers import TextStreamer
_ = model.generate(
    **tokenizer(text, return_tensors = "pt").to("cuda"),
    max_new_tokens = 1000,
    temperature = 1, top_p = 0.95, top_k = 64,
    streamer = TextStreamer(tokenizer, skip_prompt = True),
)

⚖️ License & Ownership

This model weights are a sovereign asset owned by drago-2435. Generated via Smolify.ai.