Instructions to use RMDWLLC/kaiju-coder-mlx-1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RMDWLLC/kaiju-coder-mlx-1.0 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="RMDWLLC/kaiju-coder-mlx-1.0",
	filename="kaiju-coder-mlx-1.0-q8_0.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 RMDWLLC/kaiju-coder-mlx-1.0 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 RMDWLLC/kaiju-coder-mlx-1.0:Q8_0
# Run inference directly in the terminal:
llama cli -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0
# Run inference directly in the terminal:
llama cli -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

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 RMDWLLC/kaiju-coder-mlx-1.0:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

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 RMDWLLC/kaiju-coder-mlx-1.0:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

Use Docker

docker model run hf.co/RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

LM Studio
Jan

vLLM

How to use RMDWLLC/kaiju-coder-mlx-1.0 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

Ollama
How to use RMDWLLC/kaiju-coder-mlx-1.0 with Ollama:
```
ollama run hf.co/RMDWLLC/kaiju-coder-mlx-1.0:Q8_0
```

Unsloth Studio

How to use RMDWLLC/kaiju-coder-mlx-1.0 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 RMDWLLC/kaiju-coder-mlx-1.0 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 RMDWLLC/kaiju-coder-mlx-1.0 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for RMDWLLC/kaiju-coder-mlx-1.0 to start chatting

How to use RMDWLLC/kaiju-coder-mlx-1.0 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

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": "RMDWLLC/kaiju-coder-mlx-1.0:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use RMDWLLC/kaiju-coder-mlx-1.0 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

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 RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use RMDWLLC/kaiju-coder-mlx-1.0 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

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 "RMDWLLC/kaiju-coder-mlx-1.0:Q8_0" \
  --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"

Docker Model Runner
How to use RMDWLLC/kaiju-coder-mlx-1.0 with Docker Model Runner:
```
docker model run hf.co/RMDWLLC/kaiju-coder-mlx-1.0:Q8_0
```

Lemonade

How to use RMDWLLC/kaiju-coder-mlx-1.0 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RMDWLLC/kaiju-coder-mlx-1.0:Q8_0

Run and chat with the model

lemonade run user.kaiju-coder-mlx-1.0-Q8_0

List all available models

lemonade list

kaiju-coder-mlx-1.0 / NOTICE

restokes92

Kaiju-Coder MLX 1.0: model card, license, notice, Modelfile

f2f4b95 verified about 1 month ago

Raw

History Blame Contribute Delete

2.69 kB

	Kaiju-Coder MLX 1.0 by Kiyomi
	=============================

	This product includes a fine-tune of and derivative work from a third-party
	base model, redistributed under the Apache License, Version 2.0. The full
	license text is in the accompanying LICENSE file.

	-------------------------------------------------------------------------------
	Base model
	-------------------------------------------------------------------------------

	Qwen3.6-35B-A3B
	Copyright 2026 Alibaba Cloud
	Licensed under the Apache License, Version 2.0.

	The base model is available from the Qwen team at Alibaba Cloud
	(Hugging Face repo: Qwen/Qwen3.6-35B-A3B). Architecture id: qwen3_5_moe;
	35.9B total parameters with roughly 3B active per token (mixture-of-experts).

	-------------------------------------------------------------------------------
	Modifications made in this work
	-------------------------------------------------------------------------------

	This work MODIFIED the base model. Specifically:

	- A LoRA fine-tune was trained on top of the base model and fused into the
	released weights. The fine-tune data is RMDW/Kiyomi-owned deterministic
	output for a business-niche builder use case (websites, Stripe, invoices,
	leads, CRM/intake, automations).
	- The redistributed GGUF is a TEXT-ONLY derivative. The base Qwen3.6-35B-A3B
	is a vision-language model; the vision pathway is stripped in this GGUF.
	This work does not provide and does not advertise vision capabilities.
	- Tokenizer, chat-template, and serving configuration were adapted for the
	GGUF/Ollama/llama.cpp local-serving path.

	These modifications were made by Richard Echols / RMDW. As required by the
	Apache License, Version 2.0, Section 4(b), the changed files carry notices
	stating that they were changed.

	-------------------------------------------------------------------------------
	Additions and fine-tune by this work
	-------------------------------------------------------------------------------

	Kaiju-Coder MLX additions, fine-tune weights, training and packaging scripts,
	model card, and documentation
	Copyright 2026 Richard Echols / RMDW
	Licensed under the Apache License, Version 2.0.

	-------------------------------------------------------------------------------
	Attribution and endorsement
	-------------------------------------------------------------------------------

	This is an independent fine-tune. Alibaba Cloud and the Qwen team do not
	endorse, sponsor, or support this work, and nothing in this distribution
	should be read as implying such endorsement. "Qwen" and "Alibaba Cloud" are
	used only to describe the origin of the base model, as permitted by Section 6
	of the Apache License, Version 2.0.