Instructions to use iBossonline/RegesCore-1.0-9B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use iBossonline/RegesCore-1.0-9B-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf iBossonline/RegesCore-1.0-9B-GGUF: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 iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf iBossonline/RegesCore-1.0-9B-GGUF: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 iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use iBossonline/RegesCore-1.0-9B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "iBossonline/RegesCore-1.0-9B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "iBossonline/RegesCore-1.0-9B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Ollama
How to use iBossonline/RegesCore-1.0-9B-GGUF with Ollama:
```
ollama run hf.co/iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M
```
Unsloth Desktop

How to use iBossonline/RegesCore-1.0-9B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use iBossonline/RegesCore-1.0-9B-GGUF with Docker Model Runner:
```
docker model run hf.co/iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M
```

Lemonade

How to use iBossonline/RegesCore-1.0-9B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.RegesCore-1.0-9B-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use iBossonline/RegesCore-1.0-9B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf iBossonline/RegesCore-1.0-9B-GGUF: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 iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use iBossonline/RegesCore-1.0-9B-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

RegesCore-1.0-9B

RegesCore's packaging of Ornith-1.0-9B, the 9-billion-parameter agentic coding model from the DeepReinforce Team. The weights are the official Ornith GGUF files, unchanged; RegesCore adds a system prompt and setup notes, so it runs well on modest hardware with llama.cpp, LM Studio, Ollama or any OpenAI-compatible server, and inside the Reges apps.

Built by Like A King Inc. (New York), the company behind RegesCore (private, local-first AI that runs on your machine or inside your building) and Reges Code (a free, local-first AI code editor).


Model	RegesCore-1.0-9B
Weights	deepreinforce-ai/Ornith-1.0-9B (MIT), not retrained
GGUF files	Identical to Ornith-1.0-9B-GGUF, renamed. The SHA-256 of each file matches the original
RegesCore adds	The system prompt in `system-prompts/`, this run guide
Reasoning	Writes a `<think>...</think>` block before its answer
Language	English
License	MIT (see License and attribution)

What it is

RegesCore-1.0-9B is the small sibling of RegesCore-1.0-35B: the same idea, sized for 8 to 12 GB GPUs and laptops. Its focus on systems engineering, security work and agentic coding comes from the system prompt in this repository, not from retraining: version 1.0 has not been fine-tuned. A fine-tuned release will ship under a new version number with its own card.

The system prompt asks the model to label each claim as fact, inference or unknown, and to verify instead of guessing. Keep it attached; without it you get a more generic assistant.

Intended use

Authorized security work: penetration-test planning, vulnerability assessment, reverse-engineering help, log and incident triage. You must have permission to test what you point it at.
Systems and infrastructure engineering: Linux, Docker, self-hosted deployment, databases, networking, shell automation.
Lua scripting and code review.
Agentic coding on small machines, with a harness that can parse tool calls.
Private local deployments where prompts and code must not leave the machine.

Not intended for

Unauthorized access to systems, malware development, or anything unlawful.
Safety-critical decisions or control of life-safety systems (fire alarm, access control, industrial control). Use it to read, explain and draft; people decide and act.
Legal, medical or financial advice.
Anything that needs verified accuracy without a human check. Treat every answer as a draft.

Pick a file

Sizes are from the file listing on 2026-10-11 (decimal GB). You need roughly the file size in memory plus room for context.

File	Size	Notes
`RegesCore-1.0-9B-Q4_K_M.gguf`	5.63 GB	Recommended default; fits 8 GB GPUs
`RegesCore-1.0-9B-Q5_K_M.gguf`	6.47 GB	Higher fidelity on 8 GB GPUs
`RegesCore-1.0-9B-Q6_K.gguf`	7.36 GB	10 to 12 GB GPUs
`RegesCore-1.0-9B-Q8_0.gguf`	9.53 GB	12 GB GPUs and up
`RegesCore-1.0-9B-bf16.gguf`	17.92 GB	Full precision reference

How to run it

llama.cpp (OpenAI-compatible server)

./llama-server -m RegesCore-1.0-9B-Q4_K_M.gguf --jinja -c 8192 --port 8000

This serves an OpenAI-compatible API at http://localhost:8000/v1. Send the prompt from system-prompts/regescore-system-prompt-fable5.md as the system message.

LM Studio

Search LM Studio's model browser for iBossonline/RegesCore-1.0-9B-GGUF and download a file.
Load it and set the context length (start at 8,192).
Paste the system prompt from system-prompts/ into the system prompt box.
Start the local server (by default http://127.0.0.1:1234/v1).

Ollama

ollama run hf.co/iBossonline/RegesCore-1.0-9B-GGUF:Q4_K_M

Then set the system prompt with /set system followed by the prompt text, or put it in a Modelfile SYSTEM block.

Suggested settings

Setting	Value
temperature	0.4 to 0.6 (0.3 to 0.4 for tool use)
top_p	0.95
context	8K to start; more if you have memory

Limitations

It is Ornith-1.0-9B with a RegesCore system prompt, not a new or retrained model. Its strengths and weaknesses are the base model's. For benchmark results see the Ornith-1.0-9B card; no RegesCore-specific results are published.
A 9B model makes more mistakes than the 35B, and it can sound sure while being wrong. Verify commands, code and findings before you run or report them.
No vision projector is shipped in this repository.
English only has been checked.

Built for Reges

RegesCore is Like A King Inc.'s private, local-first AI platform. The Reges apps work with any model you already run; this one is a convenient default for smaller machines.

Reges: a desktop AI assistant for Windows that runs on your machine, plus Reges for Business, a private AI front desk and operations assistant installed on hardware in your building.
Reges Code: a free (while in beta) local-first AI code editor for Windows that works with LM Studio, Ollama, llama.cpp and other local servers.
Source and tools: github.com/iboss21.
Contact: use the contact form on regescore.pro.

License and attribution

Released under the MIT License, the same license as the base model.

The model weights are Ornith-1.0-9B by the DeepReinforce Team (deepreinforce-ai/Ornith-1.0-9B, MIT), redistributed unchanged from their official GGUF release; only the file names differ. The Ornith architecture builds on the Qwen3.5 family. Every answer this model gives is their work. See the Ornith repository for the original MIT notice.

Ethical use

This model knows a lot about security tooling and reverse engineering. The RegesCore system prompt points it at defensive engineering, authorized testing and systems design. Using it against systems you do not own, or do not have written permission to test, is a crime in most jurisdictions. You are responsible for how you use it. The authors give no warranty and accept no liability for misuse.

Citation

If you use this model, cite Ornith-1.0-9B, the model itself.

@misc{ornith-9b,
  title  = {{Ornith-1.0}: Agentic Coding, Open to All},
  author = {{DeepReinforce Team}},
  year   = {2026},
  url    = {https://deep-reinforce.com/ornith_1_0.html}
}

Downloads last month: 679

GGUF

Model size

9B params

Architecture

qwen35

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for iBossonline/RegesCore-1.0-9B-GGUF

Base model

ornith-ai/Ornith-1.0-9B

Quantized

(101)

this model