Instructions to use ProSeAI/ProSe-Ohio-Legal-Expert with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use ProSeAI/ProSe-Ohio-Legal-Expert with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # if on a CUDA device, also pip install mlx[cuda] # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("ProSeAI/ProSe-Ohio-Legal-Expert") prompt = "Once upon a time in" text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Unsloth Studio
How to use ProSeAI/ProSe-Ohio-Legal-Expert with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for ProSeAI/ProSe-Ohio-Legal-Expert to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for ProSeAI/ProSe-Ohio-Legal-Expert to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for ProSeAI/ProSe-Ohio-Legal-Expert to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="ProSeAI/ProSe-Ohio-Legal-Expert", max_seq_length=2048, ) - MLX LM
How to use ProSeAI/ProSe-Ohio-Legal-Expert with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "ProSeAI/ProSe-Ohio-Legal-Expert" --prompt "Once upon a time"
ProSe-Ohio-Legal-Expert
One father. Two heartbreaks. Free AI so no parent faces the system alone.
A LoRA adapter fine-tuned on Gemma 4 12B for Ohio family law, Hamilton County Juvenile Court procedures, and pro se litigation support.
Description
This model is a LoRA adapter trained on 67 Hamilton County Juvenile Court forms (HCJF) covering custody, shared parenting, paternity, support, appeals, power of attorney, and more. It provides accurate citations to the Ohio Revised Code and local court rules, helping pro se litigants understand court forms, filing procedures, and legal standards.
Trained by: Rodney D. Mullins, pro se litigant in F25-000196Z Purpose: Level the playing field for parents who can't afford $500/hr attorneys
Training
| Detail | Value |
|---|---|
| Base model | mlx-community/gemma-4-12B-it-4bit (Gemma 4 12B, 4-bit quantized) |
| Technique | LoRA (rank 8, 16 layers) via MLX on Apple Silicon |
| Dataset | Roderick3rd/hcjf-legal-forms β 67 HCJF forms in ChatML format |
| Hardware | Apple M4 Pro (48GB unified memory) |
| Training time | ~50 minutes (170 iterations) |
| Final loss | 0.560 (train), 1.591 (validation) |
| Trainable params | 10.93M / 11,907M total (0.092%) |
Usage
With MLX (Apple Silicon)
from mlx_lm import load, generate
# Load base model + adapter
model, tokenizer = load(
"mlx-community/gemma-4-12B-it-4bit",
adapter_path="ProSeAI/ProSe-Ohio-Legal-Expert"
)
# Generate
messages = [
{"role": "system", "content": "You are an Ohio family law expert specializing in Hamilton County Juvenile Court procedures. You provide accurate citations to the Ohio Revised Code and local court rules."},
{"role": "user", "content": "What forms do I need to file for custody in Hamilton County Juvenile Court?"},
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
response = generate(model, tokenizer, prompt, max_tokens=512, temperature=0.3)
print(response)
With Ollama
# Coming soon β will be available as:
# ollama pull hf.co/ProSeAI/ProSe-Ohio-Legal-Expert
Capabilities
- Ohio family law β ORC 3109.04 (custody), ORC 3111 (paternity), local juvenile court rules
- Form assistance β which form for which situation, how to complete each field
- Pro se guidance β filing procedures, deadlines, court protocols
- Best interest factors β Ohio's 10-factor analysis for custody determinations
- GAL procedures β Guardian ad Litem reports, motions, questionnaires
Limitations
- This is a LoRA adapter, not a standalone model. It requires the Gemma 4 12B base model.
- Trained on 67 HCJF forms β strong on Hamilton County procedure, weaker on other Ohio counties
- Not a substitute for licensed legal counsel
- Always verify citations against current Ohio law
Links
License
Apache 2.0 β free for any use, commercial or otherwise.
Quantized
Model tree for ProSeAI/ProSe-Ohio-Legal-Expert
Base model
mlx-community/gemma-4-12B-it-4bit