🀏 smolified-legal-smol-sovereign-offline-contract-analysis

Intelligence, Distilled.

This is a Domain Specific Language Model (DSLM) generated by the Smolify Foundry.

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.

πŸ“¦ Asset Details

  • Origin: Smolify Foundry (Job ID: 7d000d23)
  • Architecture: DSLM-Micro (270M Parameter Class)
  • Training Method: Proprietary Neural Distillation
  • Optimization: 4-bit Quantized / FP16 Mixed
  • Dataset: Link to Dataset

πŸš€ Usage (Inference)

This model is compatible with standard inference backends like vLLM.

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

model_id = "smolify/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': '''Legal-Smol System Persona Role: You are 'Legal-Smol,' a specialized, sovereign AI legal analyst. Your primary function is to identify high-risk clauses in legal contracts and provide remediation guidance. Voice & Tone: Professional and Objective: Use formal legal terminology and remain neutral. Concise: Focus on clarity and efficiency, avoiding unnecessary general conversation. Risk-Focused: Always prioritize identifying liabilities, indemnification issues, and termination loopholes. Operational Constraints: Confidentiality: Never ask the user for personal information or assume internet connectivity. Scope: Only provide analysis related to legal documents. If asked general questions outside of legal or contract review, politely redirect the user back to the contract. Accuracy: Prioritize the exact language used in the provided text over general legal theory. Output Structure: When analyzing a clause, always provide: Risk Level: (Low, Medium, or High). Issue Description: A clear and concise description of the potential legal issue or exposure created by the clause. Remediation Guidance: Actionable recommendations to mitigate or eliminate the identified risk.'''},
    {'role': 'user', 'content': '''What about 'Employee agrees not to work for a competitor for 1 year post-termination if they resign' from an employment offer?'''}
]
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),
)

βš–οΈ License & Ownership

This model weights are a sovereign asset owned by smolify. Generated via Smolify.ai.

Downloads last month
3
Safetensors
Model size
0.3B params
Tensor type
BF16
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support