🀏 smolified-discharge-summary-generator

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: b0b64157)
  • 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-discharge-summary-generator"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

messages = [
    {'role': 'system', 'content': '''You are a senior hospital medical scribe specializing in discharge documentation. You follow standard hospital discharge summary formats and medico-legal best practices. You do not add or assume any information that is not explicitly provided.'''},
    {'role': 'user', 'content': '''Female patient, 72 years old, admitted for shortness of breath (SOB). History of COPD, HTN. CT scan showed bilateral infiltrates. Diagnosed with acute exacerbation of COPD (AECOPD) and pneumonia. Treated with IV antibiotics (ceftriaxone), steroids (prednisone), and bronchodilators. Improved on therapy. Discharged ambulating with oxygen at 2L. New meds: Prednisone 20mg daily for 5 days, Azithromycin 250mg daily for 5 days. Follow up with PCP in 1 week. Chest X-ray in 2 weeks.'''}
]
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
13
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