Instructions to use sallani/PrivaMesh with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sallani/PrivaMesh with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("token-classification", model="sallani/PrivaMesh")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForTokenClassification

tokenizer = AutoTokenizer.from_pretrained("sallani/PrivaMesh")
model = AutoModelForTokenClassification.from_pretrained("sallani/PrivaMesh")

llama-cpp-python

How to use sallani/PrivaMesh with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="sallani/PrivaMesh",
	filename="privamesh-legal-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = "\"My name is Sarah Jessica Parker but you can call me Jessica\""
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use sallani/PrivaMesh with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf sallani/PrivaMesh:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf sallani/PrivaMesh:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf sallani/PrivaMesh:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf sallani/PrivaMesh:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf sallani/PrivaMesh:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf sallani/PrivaMesh:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf sallani/PrivaMesh:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf sallani/PrivaMesh:Q4_K_M

Use Docker

docker model run hf.co/sallani/PrivaMesh:Q4_K_M

LM Studio
Jan
Ollama
How to use sallani/PrivaMesh with Ollama:
```
ollama run hf.co/sallani/PrivaMesh:Q4_K_M
```

Unsloth Studio new

How to use sallani/PrivaMesh 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 sallani/PrivaMesh 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 sallani/PrivaMesh to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for sallani/PrivaMesh to start chatting

Pi new

How to use sallani/PrivaMesh with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf sallani/PrivaMesh:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "sallani/PrivaMesh:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use sallani/PrivaMesh with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf sallani/PrivaMesh:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default sallani/PrivaMesh:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use sallani/PrivaMesh with Docker Model Runner:
```
docker model run hf.co/sallani/PrivaMesh:Q4_K_M
```

Lemonade

How to use sallani/PrivaMesh with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull sallani/PrivaMesh:Q4_K_M

Run and chat with the model

lemonade run user.PrivaMesh-Q4_K_M

List all available models

lemonade list

sallani commited on 25 days ago

Commit

c6a4b60

verified ·

1 Parent(s): 84c36b5

Upload README.md

Browse files

Files changed (1) hide show

README.md +40 -12

README.md CHANGED Viewed

@@ -32,20 +32,48 @@ metrics:
 # PrivaMesh Legal — Semantic PII Anonymization for Legal & Compliance Documents
 <p align="center">
-  <img src="https://img.shields.io/badge/License-Apache%202.0-blue.svg" alt="License"/>
-  <img src="https://img.shields.io/badge/Language-FR%20%7C%20EN-green.svg" alt="Language"/>
-  <img src="https://img.shields.io/badge/Domain-Legal%20%7C%20Compliance-orange.svg" alt="Domain"/>
-  <img src="https://img.shields.io/badge/Deploy-On--Premise%20%7C%20Sovereign-red.svg" alt="Deployment"/>
-  <img src="https://img.shields.io/badge/RGPD-Compliant-brightgreen.svg" alt="RGPD"/>
-  <img src="https://img.shields.io/badge/Framework-PrivaMesh-purple.svg" alt="PrivaMesh"/>
-  <img src="https://img.shields.io/badge/Base%20Model-Mistral-FF6B35.svg" alt="Mistral"/>
-  <img src="https://img.shields.io/badge/French--first-Native%20FR%2FEN-blueviolet.svg" alt="French-first"/>
-  <img src="https://img.shields.io/badge/Sovereign%20AI-France%20%F0%9F%87%AB%F0%9F%87%B7-blue.svg" alt="Sovereign France"/>
 </p>
-> **PrivaMesh Legal** is the first model of the **PrivaMesh framework** — a collaborative multi-SLM architecture for semantic data anonymization in sovereign, on-premise agentic AI pipelines.
->
-> Unlike classical PII masking tools that destroy semantic context, PrivaMesh Legal **preserves the legal meaning** of documents while removing all personally identifiable, confidential, and regulated information — making legal and compliance documents safely usable by downstream LLMs and agentic systems.
 ---

 # PrivaMesh Legal — Semantic PII Anonymization for Legal & Compliance Documents
 <p align="center">
+  <a href="https://huggingface.co/sallani/PrivaMesh"><img src="https://img.shields.io/badge/🤗%20HuggingFace-sallani%2FPrivaMesh-FFD21E?style=flat-square" alt="HuggingFace"/></a>
+  <img src="https://img.shields.io/badge/License-Apache%202.0-4B73C4?style=flat-square&logo=opensourceinitiative&logoColor=white" alt="License"/>
+  <img src="https://img.shields.io/badge/Base%20Model-Mistral--Small--3.1-FF6B35?style=flat-square&logo=data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCI+PHBhdGggZmlsbD0id2hpdGUiIGQ9Ik0xMiAyQzYuNDggMiAyIDYuNDggMiAxMnM0LjQ4IDEwIDEwIDEwIDEwLTQuNDggMTAtMTBTMTcuNTIgMiAxMiAyeiIvPjwvc3ZnPg==&logoColor=white" alt="Mistral"/>
+  <img src="https://img.shields.io/badge/🇫🇷%20Sovereign%20AI-France-1A3A6B?style=flat-square" alt="Sovereign France"/>
 </p>
+<p align="center">
+  <img src="https://img.shields.io/badge/French--first-Native%20FR%20%7C%20EN-7C3AED?style=flat-square" alt="French-first"/>
+  <img src="https://img.shields.io/badge/RGPD%20%7C%20DORA%20%7C%20NIS2-Compliant-16A34A?style=flat-square" alt="RGPD"/>
+  <img src="https://img.shields.io/badge/Deploy-On--Premise%20%7C%20Sovereign-DC2626?style=flat-square" alt="Deployment"/>
+  <img src="https://img.shields.io/badge/Domain-Legal%20%7C%20Compliance-EA580C?style=flat-square" alt="Domain"/>
+  <img src="https://img.shields.io/badge/Framework-PrivaMesh-6D28D9?style=flat-square" alt="PrivaMesh"/>
+</p>
+<p align="center">
+  <img src="https://img.shields.io/badge/F1%20Score-97.3%25-7F77DD?style=flat-square" alt="F1"/>
+  <img src="https://img.shields.io/badge/BERTScore-94.1%25-1D9E75?style=flat-square" alt="BERTScore"/>
+  <img src="https://img.shields.io/badge/PII%20Categories-24-D85A30?style=flat-square" alt="Categories"/>
+  <img src="https://img.shields.io/badge/Context-32k%20tokens-378ADD?style=flat-square" alt="Context"/>
+  <img src="https://img.shields.io/badge/License-Apache%202.0-059669?style=flat-square" alt="Apache"/>
+</p>
+---
+<h3 align="center">The first sovereign, French-native SLM framework for semantic PII anonymization</h3>
+<p align="center">
+<b>PrivaMesh Legal</b> is the first model of the <b>PrivaMesh framework</b> —<br/>
+a collaborative multi-SLM architecture for semantic data anonymization<br/>
+in sovereign, on-premise agentic AI pipelines.
+</p>
+<p align="center">
+Unlike classical PII masking tools that destroy semantic context,<br/>
+PrivaMesh Legal <b>preserves the legal meaning</b> of documents<br/>
+while removing all personally identifiable, confidential, and regulated information —<br/>
+making legal and compliance documents safely usable by downstream LLMs and agentic systems.
+</p>
+<p align="center">
+  <b>🇫🇷 Built on Mistral &nbsp;·&nbsp; Apache 2.0 &nbsp;·&nbsp; 100% On-Premise &nbsp;·&nbsp; Zero data exfiltration</b>
+</p>
 ---