nexora-vector-v0.1/README.md

---
license: apache-2.0
base_model:
- Qwen/Qwen3-4B
tags:
- nexora
- chat
- qwen3
- conversational
language:
- en
pipeline_tag: text-generation
library_name: transformers
---
<p align="center">
  <img src="https://huggingface.co/ArkAiLab-Adl/nexora-vector-v0.1/resolve/main/assets/nexora-vector.png" alt="Nexora-Vector"/>
</p>

# Nexora-Vector-v0.1

<p align="center">
  <img src="https://img.shields.io/badge/status-beta-orange" alt="Status: Beta"/>
  <img src="https://img.shields.io/badge/license-Apache%202.0-blue" alt="License: Apache 2.0"/>
  <img src="https://img.shields.io/badge/base_model-Qwen3--4B-blueviolet" alt="Base Model"/>
  <img src="https://img.shields.io/badge/output-SVG-green" alt="Output: SVG"/>
</p>

> **Nexora-Vector-v0.1** is an experimental text-to-vector model that generates structured SVG graphics from natural language prompts. This is the inaugural release of the Nexora Vector series, intended for research, prototyping, and early-stage development workflows.

---

## ⚠️ Update Notice

An issue was identified in the initial release where an incorrect base model was uploaded.  
This has now been **fully corrected**, and the current version is properly based on **Qwen3-4B**.

Users are advised to **re-download the latest version** to ensure correct behavior and performance.

---

## Table of Contents

- [Overview](#overview)
- [Model Details](#model-details)
- [Capabilities](#capabilities)
- [Limitations](#limitations)
- [Intended Use](#intended-use)
- [Architecture & Training](#architecture--training)
- [Usage Recommendations](#usage-recommendations)
- [Quantized Versions](#quantized-versions)
- [Evaluation](#evaluation)
- [Risks & Considerations](#risks--considerations)
- [Future Work](#future-work)
- [Community & Support](#community--support)
- [License](#license)
- [Acknowledgements](#acknowledgements)

---

## Overview

Nexora-Vector-v0.1 is a supervised fine-tuned language model built on top of **Qwen3-4B**, adapted specifically to generate structured vector graphics in SVG format from natural language instructions.

This release is in **beta** and is scoped to research, experimentation, and early-stage design tooling. All outputs should be validated before use in any downstream pipeline.

---

## Capabilities

Nexora-Vector-v0.1 is designed to translate textual instructions into structured SVG code. The model is best suited for:

- Generating SVG markup for simple vector graphics
- Producing geometric shapes and basic illustrations
- Creating lightweight icons and minimal design assets
- Supporting rapid prototyping in vector-based design workflows

> **Tip:** The model performs best with concise, clearly scoped prompts focused on simple visual compositions.

---

## Limitations

This is an early-stage beta release. Users should be aware of the following constraints before integrating the model:

- **High hallucination rate** — outputs may be invalid or non-renderable SVG
- **Limited generalization** — the small dataset size affects output consistency
- **Weak complex scene handling** — highly detailed or multi-element prompts may produce poor results
- **Manual correction required** — outputs should be validated and post-processed before use
- **Not production-ready** — not suitable for safety-critical or automated pipelines

---

## Intended Use

### ✅ Supported Use Cases

- Academic and applied research in text-to-vector generation
- Experimental AI-assisted design systems
- Educational exploration of structured output generation
- Lightweight SVG prototyping and ideation

### ❌ Out-of-Scope Use Cases

- Production-grade or commercial vector asset pipelines
- High-precision design deliverables without human validation
- Automated systems where SVG correctness is required without manual review

---

## Architecture & Training

The model is built on **Qwen3-4B** and fine-tuned using supervised learning to improve structured SVG output generation.

### Training Configuration

| Parameter | Details |
|---|---|
| **Fine-tuning Method** | Supervised Fine-Tuning (SFT) |
| **Dataset Composition** | Curated prompt–SVG pairs |
| **Dataset Size** | ~1,500 samples |
| **Training Objective** | Structured output generation for SVG formats |

> **Note:** The relatively small dataset size may result in instability and limited generalization across diverse prompts. Improved dataset coverage is planned for future versions.

---

## Usage Recommendations

To get the best results from Nexora-Vector-v0.1:

1. **Keep prompts simple and specific** — avoid multi-scene or highly complex compositions
2. **Validate all SVG outputs** before rendering or integrating into any pipeline
3. **Post-process outputs** to correct syntax or structural issues
4. **Use iterative prompting** — refining prompts across multiple turns often yields better results
5. **Expect imperfections** — this is a beta model; treat outputs as drafts, not finals

---

## Quantized Versions

Official quantized releases are available via **[Open4bits](https://huggingface.co/Open4bits)** — the dedicated quantization project under **ArkAiLabs** — for efficient local inference across different hardware platforms:

| Version | Format | Link |
|---|---|---|
| **GGUF** (Q2_K / Q4_K_M / Q6_K / Q8_0) | GGUF |  [Open4bits/nexora-vector-v0.1-GGUF](https://huggingface.co/Open4bits/nexora-vector-v0.1-GGUF) |
| **MLX 4-Bit** (Apple Silicon) | MLX | [Open4bits/nexora-vector-v0.1-mlx-4Bit](https://huggingface.co/Open4bits/nexora-vector-v0.1-mlx-4Bit) |

- Use the **GGUF** version for local inference on Windows, Linux, or macOS with tools like `llama.cpp`, Ollama, or LM Studio.
- Use the **MLX** version for optimized inference on Apple Silicon (M1/M2/M3/M4) via the MLX framework.

---

## Evaluation

Nexora-Vector-v0.1 has not yet undergone formal benchmark evaluation. Current assessment is qualitative, based on manual testing of SVG generation tasks.

Planned evaluation metrics for future releases include:

| Metric | Description |
|---|---|
| **SVG Validity Rate** | Percentage of outputs that are parseable, valid SVG |
| **Structural Correctness** | Adherence to SVG schema and element hierarchy |
| **Prompt Adherence** | Alignment between user intent and generated output |
| **Visual Consistency** | Stability of outputs across similar prompts |

---

## Risks & Considerations

Developers integrating Nexora-Vector-v0.1 should account for the following risks:

- Generation of malformed or non-functional SVG code
- Inconsistent instruction following across prompt variations
- Unpredictable outputs due to limited training data coverage

**Recommendation:** Implement downstream validation layers and SVG syntax checking before any rendering or integration.

---

## Future Work

The following improvements are planned for upcoming versions of the Nexora Vector series:

- [ ] Expanded and more diverse training dataset
- [ ] Improved SVG syntax correctness and validity rates
- [ ] Reduced hallucination rates
- [ ] Enhanced natural language understanding for complex prompts
- [ ] Support for richer vector compositions and multi-element scenes
- [ ] Formal benchmark evaluation suite

---

## Community & Support

Join the community for updates and discussion:

💬 **[Join our Discord Server](https://discord.gg/mwdrgYbzuG)**

---

## License

This model is released under the **Apache License 2.0**.

You may use, modify, and distribute this model in accordance with the terms of the Apache 2.0 license. See the [LICENSE](./LICENSE) file for full details, or refer to the [official Apache 2.0 license text](https://www.apache.org/licenses/LICENSE-2.0).

---

## Acknowledgements

Nexora-Vector-v0.1 is built upon **[Qwen3-4B](https://huggingface.co/Qwen/Qwen3-4B)** by the Qwen team. We thank the open-source AI community for their continued contributions that make projects like this possible.

---

## About Nexora

**Nexora** is an experimental AI initiative under **ArkAiLabs**, focused on building lightweight, practical, and creative AI systems for real-world applications. The Nexora Vector series represents our exploration into AI-assisted vector graphics generation.