Instructions to use 0sz1/IT-Auditor-8B-PRO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use 0sz1/IT-Auditor-8B-PRO with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="0sz1/IT-Auditor-8B-PRO") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("0sz1/IT-Auditor-8B-PRO", dtype="auto") - llama-cpp-python
How to use 0sz1/IT-Auditor-8B-PRO with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="0sz1/IT-Auditor-8B-PRO", filename="Auditor_8B_PRO_Q8.gguf", )
llm.create_chat_completion( messages = [ { "role": "user", "content": "What is the capital of France?" } ] ) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use 0sz1/IT-Auditor-8B-PRO with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf 0sz1/IT-Auditor-8B-PRO # Run inference directly in the terminal: llama-cli -hf 0sz1/IT-Auditor-8B-PRO
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf 0sz1/IT-Auditor-8B-PRO # Run inference directly in the terminal: llama-cli -hf 0sz1/IT-Auditor-8B-PRO
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 0sz1/IT-Auditor-8B-PRO # Run inference directly in the terminal: ./llama-cli -hf 0sz1/IT-Auditor-8B-PRO
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 0sz1/IT-Auditor-8B-PRO # Run inference directly in the terminal: ./build/bin/llama-cli -hf 0sz1/IT-Auditor-8B-PRO
Use Docker
docker model run hf.co/0sz1/IT-Auditor-8B-PRO
- LM Studio
- Jan
- vLLM
How to use 0sz1/IT-Auditor-8B-PRO with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "0sz1/IT-Auditor-8B-PRO" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "0sz1/IT-Auditor-8B-PRO", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/0sz1/IT-Auditor-8B-PRO
- SGLang
How to use 0sz1/IT-Auditor-8B-PRO with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "0sz1/IT-Auditor-8B-PRO" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "0sz1/IT-Auditor-8B-PRO", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "0sz1/IT-Auditor-8B-PRO" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "0sz1/IT-Auditor-8B-PRO", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Ollama
How to use 0sz1/IT-Auditor-8B-PRO with Ollama:
ollama run hf.co/0sz1/IT-Auditor-8B-PRO
- Unsloth Studio
How to use 0sz1/IT-Auditor-8B-PRO 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 0sz1/IT-Auditor-8B-PRO 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 0sz1/IT-Auditor-8B-PRO to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for 0sz1/IT-Auditor-8B-PRO to start chatting
- Pi
How to use 0sz1/IT-Auditor-8B-PRO with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf 0sz1/IT-Auditor-8B-PRO
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": "0sz1/IT-Auditor-8B-PRO" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use 0sz1/IT-Auditor-8B-PRO with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf 0sz1/IT-Auditor-8B-PRO
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 0sz1/IT-Auditor-8B-PRO
Run Hermes
hermes
- Docker Model Runner
How to use 0sz1/IT-Auditor-8B-PRO with Docker Model Runner:
docker model run hf.co/0sz1/IT-Auditor-8B-PRO
- Lemonade
How to use 0sz1/IT-Auditor-8B-PRO with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull 0sz1/IT-Auditor-8B-PRO
Run and chat with the model
lemonade run user.IT-Auditor-8B-PRO-{{QUANT_TAG}}List all available models
lemonade list
LIVE DEMO AVAILABLE HERE Test the model's intelligence in your browser before you buy.
IT-Auditor-8B-PRO: Specialized 8B Model for Contract Auditing
IT-Auditor-8B-PRO is a high-precision, industrial-grade Large Language Model designed specifically for freelancers, indie developers, and small IT firms. It doesn't just read your contractsโit provides structured, risk-focused analysis suitable for legal review.
Why This Model?
Most AI models are "polite generalists." They are trained to be nice, not protective. IT-Auditor-8B-PRO is different:
- Teacher Core: Knowledge distilled from Llama 3.1 70B experts.
- Extreme Training: Forged on an NVIDIA RTX 5090 (Blackwell) with 16-bit precision and a high LoRA rank (128).
- Deep Grounding: Trained on 10,000+ expert-annotated samples, including real-world predatory filings from SEC EDGAR (Apple, NVIDIA, Palantir).
- Surgical Precision: Achieved a training loss of ~0.23, ensuring near-perfect logical alignment with professional legal reasoning.
Core Capabilities
This model is fine-tuned to identify and mitigate:
- Intellectual Property Traps: Detects when rights transfer before you get paid.
- Liability Ambiguities: Flags "indirect/consequential" damages without caps.
- Unilateral Termination: Spots 1-hour/24-hour notice periods and power imbalances.
- Automatic Redlining: Every audit includes a SUGGESTED REDRAFTโlegally sound text you can copy-paste to defend your interests.
Proof of Performance: The "Acid Test"
Input: "The Company reserves the right to withhold payment if the work is deemed unsatisfactory in its sole discretion... Contractor is strictly prohibited from providing similar services to any competitor worldwide for 5 years."*
Auditor PRO:
Proof of Performance: The "Open Source" Trap
Input: The Contractor represents and warrants that the Work Product does not infringe upon any third-party rights. Contractor shall indemnify and defend the Company at its own expense against any intellectual property claims arising from the use of the Work Product, including claims related to any open-source components, third-party libraries, or frameworks integrated into the software.
Auditor PRO:
Proof of Performance: The "Bankruptcy" Trap
Input: Contractor agrees to be liable for all damages resulting from any system downtime, bugs, or errors in the delivered software. This liability includes, without limitation, compensation for the Companyโs lost profits, loss of business reputation, and any consequential or incidental damages. Contractorโs liability under this section is absolute and shall not be capped by the total fees paid under this Agreement.
Auditor PRO:
API Infrastructure & Licensing
The AetherSieve 8B PRO engine is now accessible via a dedicated high-speed inference gateway. This infrastructure is optimized for professional auditing workflows, providing the logical precision of frontier architectures with significantly lower computational overhead.
Provisioning Tiers
[Tier 1] Scout Access โ $29.00
Allocation: 1,000,000 Semantic Tokens. Use Case: Integration testing, PoC, and individual clause evaluation. Acquire License Key
[Tier 2] Advanced Pass โ $149.00
Allocation: 10,000,000 Semantic Tokens. Efficiency: 50% reduction in cost per token compared to Tier 1. Use Case: Production-scale auditing and high-volume document processing. Acquire License Key
Technical Integration
The gateway is fully OpenAI-compatible, allowing for seamless integration into existing pipelines. Endpoint: http://79.117.5.114:29322/v1 Method: POST chat/completions Balance Monitoring: http://79.117.5.114:29322/v1/status Model Identifier: Auditor_8B_PRO_F16
Disclaimer
IT-Auditor-8B-PRO is a research and decision-support tool. It DOES NOT provide legal advice and DOES NOT create an attorney-client relationship. Always verify outputs with a qualified human legal professional.
Copyright (c) 2026. All rights reserved.
- Downloads last month
- 50
We're not able to determine the quantization variants.
Model tree for 0sz1/IT-Auditor-8B-PRO
Base model
meta-llama/Llama-3.1-8B


# Gated model: Login with a HF token with gated access permission hf auth login