PEFT
Safetensors
llama4_text
legal
contract-review
redlining
autoscientist
lora
4-bit precision
bitsandbytes
Instructions to use brimbim/legal-contract-review-model-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use brimbim/legal-contract-review-model-v1 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("togethercomputer/Llama-4-Scout-17B-16E-Instruct_bnb_4bit") model = PeftModel.from_pretrained(base_model, "brimbim/legal-contract-review-model-v1") - Notebooks
- Google Colab
- Kaggle
Legal Contract Review Assistant v1
Expert legal AI assistant specialized in commercial contract review and redlining for small businesses and freelancers.
Model Details
- Developed for: Adaption Labs AutoScientist Challenge (Legal Category, Part 1)
- Base Model: meta-llama/Llama-4-Scout-17B-16
- Training Method: LoRA fine-tuning with AutoScientist
- Win Rate: 76% (Legal category)
Intended Use
- Contract clause identification and risk assessment
- Redline suggestions protecting smaller parties
- Plain English explanations of legal terms
Training Data
- Expert-annotated commercial contracts
- Synthetic redlining examples focused on small business protection
- Total ~50k high-quality instruction-output pairs
Performance
- Win Rate: 76% on hidden test set
- Strong improvement over base model in legal reasoning and practical advice
Limitations
- Not a substitute for professional legal advice
- Best used as an assistant tool for experienced users
Citation
If you use this model, please cite the Adaption Labs AutoScientist Challenge.
- Downloads last month
- 21
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support