Instructions to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("KoarAI/QuestionMark-Ultimate-0.6B-DUS")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS 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 KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M

Use Docker

docker model run hf.co/KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KoarAI/QuestionMark-Ultimate-0.6B-DUS"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KoarAI/QuestionMark-Ultimate-0.6B-DUS",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M

Ollama
How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with Ollama:
```
ollama run hf.co/KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M
```
Unsloth Desktop

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "KoarAI/QuestionMark-Ultimate-0.6B-DUS"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "KoarAI/QuestionMark-Ultimate-0.6B-DUS"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "KoarAI/QuestionMark-Ultimate-0.6B-DUS"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "KoarAI/QuestionMark-Ultimate-0.6B-DUS"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "KoarAI/QuestionMark-Ultimate-0.6B-DUS",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with Docker Model Runner:
```
docker model run hf.co/KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M
```

Lemonade

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull KoarAI/QuestionMark-Ultimate-0.6B-DUS:Q4_K_M

Run and chat with the model

lemonade run user.QuestionMark-Ultimate-0.6B-DUS-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "KoarAI/QuestionMark-Ultimate-0.6B-DUS"

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 KoarAI/QuestionMark-Ultimate-0.6B-DUS

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use KoarAI/QuestionMark-Ultimate-0.6B-DUS with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "KoarAI/QuestionMark-Ultimate-0.6B-DUS"

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 "KoarAI/QuestionMark-Ultimate-0.6B-DUS" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

QuestionMark-Ultimate-0.6B-DUS / README.md

KoarAI

Upload README.md with huggingface_hub

758db72 verified 12 days ago

preview code

Raw History Blame Contribute Delete

5.93 kB

---
language:
- ru
- en
- uk
- be
- kk
- bg
license: apache-2.0
tags:
- text-generation
- mlx
- gguf
- merge
- ties-merge
- depth-up-scaling
- dus
- orpo
- rl
- reasoning
- tool-calling
- apple-silicon
- metal
- on-device
pipeline_tag: text-generation
inference: true
base_model:
- Qwen/Qwen2.5-0.5B-Instruct
model_creator: KoarAI
model_name: QuestionMark-Ultimate-0.6B-DUS
model_type: qwen2
---
<div align="center">

# ⚡ QuestionMark-Ultimate-0.6B-DUS (32 Layers)

### *On-Device Sub-Billion Intelligence: TIES-Merged, 32-Layer DUS & Cyrillic ORPO*

[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
[![Base Model](https://img.shields.io/badge/Base-Qwen2.5--0.5B-7952b3.svg)](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct)
[![Architecture](https://img.shields.io/badge/Architecture-32L%20DUS-teal.svg)](#-model-architecture)
[![Alignment](https://img.shields.io/badge/RL-ORPO%20Aligned-darkgreen.svg)](#-key-features)
[![Tool Calling](https://img.shields.io/badge/Tool%20Calling-JSON%20Schema-crimson.svg)](#-tool-calling)

<br/>

**QuestionMark-Ultimate-0.6B-DUS** is an ultra-fast, sub-billion parameter language model developed by **KoarAI**, engineered for local on-device inference across consumer hardware (Apple Silicon, edge devices, single-board computers, and standard PCs).

The model was scaled from 24 to **32 layers** using non-destructive **Depth Up-Scaling (DUS)**, merged across three domain-specific expert checkpoints via **TIES-Merging** (pruning 35% of parameter noise and interference vectors), and post-trained with **Safe On-Device ORPO (Odds Ratio Preference Optimization)** on Apple Silicon Metal GPU for multilingual Cyrillic understanding and native structured tool calling.

</div>

---

## 🌟 Key Features

- 🧠 **Depth Up-Scaled (DUS) to 32 Layers:** Expanded reasoning capacity compared to standard 24-layer baseline architectures.
- 🔀 **TIES-Merging:** Dense parameter combination of three specialized checkpoints (`Coder-0.5B`, `Fusion-0.5B`, and `General-0.5B`) with interference vector trimming.
- 🇷🇺 **Multilingual Cyrillic Alignment:** Tuned on curated preference pairs covering Russian dialogue, grammar, and Slavic Cyrillic variations without degrading code logic.
- 🛠️ **Structured Function & Tool Calling:** Directly extracts arguments and generates valid JSON payloads (`<tool_call>{"name": ..., "arguments": ...}</tool_call>`) from conversational user prompts.
- ⚡ **Lightweight Footprint:** Requires under 700 MB of RAM / VRAM, running at sustained speeds of **50–60 tokens/second** on consumer hardware.

---

## 📊 Benchmark & Capabilities

| Benchmark Category | Task Example | Result |
| :--- | :--- | :---: |
| **Python Algorithms** | QuickSort, palindrome verification, generators | ✅ **PASS** |
| **Tool Calling** | Structured JSON function argument extraction | ✅ **PASS** |
| **Arithmetic & Step Logic** | Multi-step word math reasoning | ✅ **PASS** |
| **Russian Dialogue** | Natural conversational flow without loop degradation | ✅ **PASS** |

---

## 🚀 Quickstart & Inference

### 1. In Ollama / LM Studio / llama.cpp (GGUF)

Pre-quantized GGUF binaries are provided directly in this repository:
- **`QuestionMark-Ultimate-0.6B-DUS32L-Q8_0.gguf`** (657 MB — Recommended: full precision, accurate EOS handling)
- **`QuestionMark-Ultimate-0.6B-DUS32L-Q4_K_M.gguf`** (480 MB — Ultra-compact footprint)
```bash
llama-cli -m QuestionMark-Ultimate-0.6B-DUS32L-Q8_0.gguf \
  -p "<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nWrite a quicksort implementation in Python.<|im_end|>\n<|im_start|>assistant\n" \
  -n 512 --temp 0.2
```
### 2. In Apple MLX (Metal GPU)
```python
from mlx_lm import load, generate
model, tokenizer = load("KoarAI/QuestionMark-Ultimate-0.6B-DUS")
response = generate(
    model, 
    tokenizer, 
    prompt="<|im_start|>user\nHello! What tasks can you perform?<|im_end|>\n<|im_start|>assistant\n",
    max_tokens=256
)
print(response)
```
### 3. In Hugging Face Transformers (PyTorch / Safetensors)
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "KoarAI/QuestionMark-Ultimate-0.6B-DUS"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
prompt = "<|im_start|>user\nWrite a Python function to check if a string is a palindrome.<|im_end|>\n<|im_start|>assistant\n"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=256, temperature=0.2)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))
```
---
## 🛠️ Tool Calling Specification
The model natively adopts the standard ChatML tool-calling format:
**User Prompt:**
> *"What is the current weather in Krasnoyarsk?"*
**Assistant Output:**
```json
{
  "name": "get_current_weather",
  "arguments": {
    "city": "Красноярск"
  }
}
```
---

## ⚙️ Model Architecture Specifications

- **Base Architecture:** Qwen2.5 (Dense Decoder-Only Transformer)
- **Active Parameters:** ~600M (0.6B)
- **Layers:** **32** (Depth Up-Scaled via residual interpolation from 24)
- **Attention:** 14 query heads (2 KV heads, Grouped Query Attention)
- **Context Length:** 4096 tokens (Native 32K context window supported)
- **Training Pipeline:** TIES-Merging (Top 35% density trimming) + Safe On-Device ORPO RL

---

## 📜 Citation

If you incorporate this model, weights, or architecture in your work, please cite:

```bibtex
@misc{koarai2026questionmarkultimate,
  author = {KoarAI Studio},
  title = {QuestionMark-Ultimate-0.6B-DUS: On-Device 32-Layer LLM with Tool Calling and Cyrillic RL},
  year = {2026},
  publisher = {Hugging Face},
  howpublished = {\\url{https://huggingface.co/KoarAI/QuestionMark-Ultimate-0.6B-DUS}}
}
```

	---
	language:
	- ru
	- en
	- uk
	- be
	- kk
	- bg
	license: apache-2.0
	tags:
	- text-generation
	- mlx
	- gguf
	- merge
	- ties-merge
	- depth-up-scaling
	- dus
	- orpo
	- rl
	- reasoning
	- tool-calling
	- apple-silicon
	- metal
	- on-device
	pipeline_tag: text-generation
	inference: true
	base_model:
	- Qwen/Qwen2.5-0.5B-Instruct
	model_creator: KoarAI
	model_name: QuestionMark-Ultimate-0.6B-DUS
	model_type: qwen2
	---

	<div align="center">

	# ⚡ QuestionMark-Ultimate-0.6B-DUS (32 Layers)

	### On-Device Sub-Billion Intelligence: TIES-Merged, 32-Layer DUS & Cyrillic ORPO

	[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
	[![Base Model](https://img.shields.io/badge/Base-Qwen2.5--0.5B-7952b3.svg)](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct)
	[![Architecture](https://img.shields.io/badge/Architecture-32L%20DUS-teal.svg)](#-model-architecture)
	[![Alignment](https://img.shields.io/badge/RL-ORPO%20Aligned-darkgreen.svg)](#-key-features)
	[![Tool Calling](https://img.shields.io/badge/Tool%20Calling-JSON%20Schema-crimson.svg)](#-tool-calling)

	<br/>

	QuestionMark-Ultimate-0.6B-DUS is an ultra-fast, sub-billion parameter language model developed by KoarAI, engineered for local on-device inference across consumer hardware (Apple Silicon, edge devices, single-board computers, and standard PCs).

	The model was scaled from 24 to 32 layers using non-destructive Depth Up-Scaling (DUS), merged across three domain-specific expert checkpoints via TIES-Merging (pruning 35% of parameter noise and interference vectors), and post-trained with Safe On-Device ORPO (Odds Ratio Preference Optimization) on Apple Silicon Metal GPU for multilingual Cyrillic understanding and native structured tool calling.

	</div>

	---

	## 🌟 Key Features

	- 🧠 Depth Up-Scaled (DUS) to 32 Layers: Expanded reasoning capacity compared to standard 24-layer baseline architectures.
	- 🔀 TIES-Merging: Dense parameter combination of three specialized checkpoints (`Coder-0.5B`, `Fusion-0.5B`, and `General-0.5B`) with interference vector trimming.
	- 🇷🇺 Multilingual Cyrillic Alignment: Tuned on curated preference pairs covering Russian dialogue, grammar, and Slavic Cyrillic variations without degrading code logic.
	- 🛠️ Structured Function & Tool Calling: Directly extracts arguments and generates valid JSON payloads (`<tool_call>{"name": ..., "arguments": ...}</tool_call>`) from conversational user prompts.
	- ⚡ Lightweight Footprint: Requires under 700 MB of RAM / VRAM, running at sustained speeds of 50–60 tokens/second on consumer hardware.

	---

	## 📊 Benchmark & Capabilities

	\| Benchmark Category \| Task Example \| Result \|
	\| :--- \| :--- \| :---: \|
	\| Python Algorithms \| QuickSort, palindrome verification, generators \| ✅ PASS \|
	\| Tool Calling \| Structured JSON function argument extraction \| ✅ PASS \|
	\| Arithmetic & Step Logic \| Multi-step word math reasoning \| ✅ PASS \|
	\| Russian Dialogue \| Natural conversational flow without loop degradation \| ✅ PASS \|

	---

	## 🚀 Quickstart & Inference

	### 1. In Ollama / LM Studio / llama.cpp (GGUF)

	Pre-quantized GGUF binaries are provided directly in this repository:
	- `QuestionMark-Ultimate-0.6B-DUS32L-Q8_0.gguf` (657 MB — Recommended: full precision, accurate EOS handling)
	- `QuestionMark-Ultimate-0.6B-DUS32L-Q4_K_M.gguf` (480 MB — Ultra-compact footprint)

	```bash
	llama-cli -m QuestionMark-Ultimate-0.6B-DUS32L-Q8_0.gguf \
	-p "<\|im_start\|>system\nYou are a helpful assistant.<\|im_end\|>\n<\|im_start\|>user\nWrite a quicksort implementation in Python.<\|im_end\|>\n<\|im_start\|>assistant\n" \
	-n 512 --temp 0.2
	```

	### 2. In Apple MLX (Metal GPU)

	```python
	from mlx_lm import load, generate

	model, tokenizer = load("KoarAI/QuestionMark-Ultimate-0.6B-DUS")
	response = generate(
	model,
	tokenizer,
	prompt="<\|im_start\|>user\nHello! What tasks can you perform?<\|im_end\|>\n<\|im_start\|>assistant\n",
	max_tokens=256
	)
	print(response)
	```

	### 3. In Hugging Face Transformers (PyTorch / Safetensors)

	```python
	from transformers import AutoModelForCausalLM, AutoTokenizer

	model_id = "KoarAI/QuestionMark-Ultimate-0.6B-DUS"

	tokenizer = AutoTokenizer.from_pretrained(model_id)
	model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

	prompt = "<\|im_start\|>user\nWrite a Python function to check if a string is a palindrome.<\|im_end\|>\n<\|im_start\|>assistant\n"
	inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
	outputs = model.generate(**inputs, max_new_tokens=256, temperature=0.2)
	print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))
	```

	---

	## 🛠️ Tool Calling Specification

	The model natively adopts the standard ChatML tool-calling format:

	User Prompt:
	> "What is the current weather in Krasnoyarsk?"

	Assistant Output:
	```json
	{
	"name": "get_current_weather",
	"arguments": {
	"city": "Красноярск"
	}
	}
	```

	---

	## ⚙️ Model Architecture Specifications

	- Base Architecture: Qwen2.5 (Dense Decoder-Only Transformer)
	- Active Parameters: ~600M (0.6B)
	- Layers: 32 (Depth Up-Scaled via residual interpolation from 24)
	- Attention: 14 query heads (2 KV heads, Grouped Query Attention)
	- Context Length: 4096 tokens (Native 32K context window supported)
	- Training Pipeline: TIES-Merging (Top 35% density trimming) + Safe On-Device ORPO RL

	---

	## 📜 Citation

	If you incorporate this model, weights, or architecture in your work, please cite:

	```bibtex
	@misc{koarai2026questionmarkultimate,
	author = {KoarAI Studio},
	title = {QuestionMark-Ultimate-0.6B-DUS: On-Device 32-Layer LLM with Tool Calling and Cyrillic RL},
	year = {2026},
	publisher = {Hugging Face},
	howpublished = {\\url{https://huggingface.co/KoarAI/QuestionMark-Ultimate-0.6B-DUS}}
	}
	```