README.md

---
license: apache-2.0
language:
  - en
library_name: transformers
pipeline_tag: text-generation
tags:
  - code
  - code-generation
  - coding-assistant
  - qlora
  - unsloth
  - local-inference
  - merkium-ai
  - adamas-mini-1
model-index:
  - name: Arti Code Mini
    results: []
---

<div align="center">

<svg width="640" height="108" viewBox="0 0 640 108" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="Arti Code Mini banner">
  <rect width="640" height="108" fill="#111111"/>
  <rect x="0" y="0" width="640" height="28" fill="#1c1c1c"/>
  <circle cx="16" cy="14" r="5" fill="#4a4a4a"/>
  <circle cx="34" cy="14" r="5" fill="#4a4a4a"/>
  <circle cx="52" cy="14" r="5" fill="#4a4a4a"/>
  <text x="320" y="18" font-family="'SF Mono', Consolas, monospace" font-size="11" fill="#7a7a7a" text-anchor="middle">arti-code-mini</text>
  <text x="24" y="60" font-family="'SF Mono', Consolas, monospace" font-size="22" fill="#f2f2f2">&gt; Arti Code Mini</text>
  <text x="24" y="84" font-family="'SF Mono', Consolas, monospace" font-size="13" fill="#8a8a8a">Local Coding</text>
  <rect x="24" y="93" width="8" height="14" fill="#f2f2f2">
    <animate attributeName="opacity" values="1;1;0;0" dur="1s" repeatCount="indefinite"/>
  </rect>
</svg>

<sub>apache-2.0 &nbsp;·&nbsp; 3B params &nbsp;·&nbsp; 2,048 ctx &nbsp;·&nbsp; 6GB+ VRAM</sub>

</div>

# Arti Code Mini

Arti Code Mini is a 3B-parameter coding assistant fine-tuned by **Merkium AI** for local deployment on consumer hardware. It targets clean, well-structured code generation and multi-turn debugging assistance without a dependency on cloud inference.

<details>
<summary><b>Table of Contents</b></summary>

- [Model Details](#model-details)
- [Intended Use](#intended-use)
- [Installation](#installation)
- [Usage](#usage)
- [Hardware Requirements](#hardware-requirements)
- [Example Prompts](#example-prompts)
- [Limitations](#limitations)
- [Citation](#citation)
- [License](#license)
- [Contact](#contact)

</details>

**Quick start**

```bash
pip install transformers torch accelerate bitsandbytes
python -c "
from transformers import pipeline
pipe = pipeline('text-generation', model='Merkiumai/Arti-code-mini', device_map='auto')
print(pipe([{'role': 'user', 'content': 'Write a function that reverses a string.'}], max_new_tokens=200)[0]['generated_text'][-1]['content'])
"
```

## Model Details

| Property | Value |
|---|---|
| Developer | Merkium AI |
| Parameters | 3 Billion |
| Fine-tuning method | QLoRA (LoRA via Unsloth) |
| Primary use case | Coding assistant |
| Context length | 2,048 tokens |
| Precision | float16 / 4-bit quantized |
| License | Apache 2.0 |
| tuned | Adamas Mini 1 |

## Intended Use

**In scope:**
- Writing and completing code across multiple languages
- Debugging and explaining existing code
- Generating functions, classes, algorithms, and scripts
- Learning support for programming concepts and best practices

**Out of scope:**
- General conversation or non-coding tasks
- Fully autonomous code generation without human review
- Production systems without independent testing and validation

## Installation

```bash
pip install transformers torch accelerate bitsandbytes
```

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "Merkiumai/Arti-code-mini"

tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto",
    trust_remote_code=True
)

messages = [
    {"role": "user", "content": "Write a Python function that checks if a number is prime."}
]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)

outputs = model.generate(
    **inputs,
    max_new_tokens=300,
    temperature=0.3,
    do_sample=True,
    pad_token_id=tokenizer.eos_token_id
)

response = tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True)
print(response)
```

Chat formatting is applied automatically via `tokenizer.apply_chat_template()`. The underlying format:

```text
### System:
You are Arti Code Mini, a helpful coding assistant created by Merkium AI.

### User:
Write a Python function that reverses a string.

### Arti:
def reverse_string(s: str) -> str:
    return s[::-1]
```

## Hardware Requirements

| Component | Minimum | Recommended |
|---|---|---|
| GPU VRAM | 6 GB | 8 GB+ |
| RAM | 8 GB | 16 GB |
| Storage | 4 GB free | 8 GB free |

## Example Prompts

- Write a Python function that checks if a string is a palindrome.
- Create a FastAPI endpoint that accepts a name and returns a greeting.
- Explain how binary search works and provide an implementation.
- Write a class for a bank account with deposit and withdraw methods.
- Write a decorator that measures the execution time of a function.
- Remove duplicates from a list while preserving order.

## Limitations

- May occasionally produce incorrect or incomplete code.
- Performs best with clear, specific, well-structured prompts.
- Not suited to tasks outside coding and software development.
- Context limited to 2,048 tokens per session.
- All generated code should be reviewed by a human before use in production.

## Citation

```bibtex
@misc{artimini2025,
  title  = {Arti Code Mini: A Lightweight Local Coding Assistant},
  author = {{Merkium AI}},
  year   = {2025},
  url    = {https://huggingface.co/Merkiumai/Arti-code-mini}
}
```

## License

Released under the [Apache 2.0](LICENSE) license.

## Contact

Model repository: [huggingface.co/Merkiumai/Arti-code-mini](https://huggingface.co/Merkiumai/Arti-code-mini)

---

<p align="center"><sub>Developed and maintained by Merkium AI</sub></p>