12 KiB
base_model, license, language, pipeline_tag, tags
| base_model | license | language | pipeline_tag | tags | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
llama3.2 |
|
text-generation |
|
Llama-Nexora-Vector-v0.1 — GGUF
This is the official GGUF quantized release of llama-nexora-vector-v0.1, published by Open4bits — the official quantization project under ArkAiLabs. Multiple quantization levels are provided to suit a wide range of hardware configurations. This is a beta release intended for research, prototyping, and early-stage development workflows only.
Table of Contents
- Overview
- The Llama-Nexora Family
- Available Quantizations
- Which Quant Should I Use?
- Model Details
- Requirements
- Installation & Usage
- Capabilities
- Limitations
- Intended Use
- Usage Recommendations
- Risks & Considerations
- Community & Support
- License
- Acknowledgements
Overview
llama-nexora-vector-v0.1-GGUF contains the official GGUF quantized versions of llama-nexora-vector-v0.1 — an experimental text-to-vector model from the Llama-Nexora family that generates structured SVG graphics from natural language prompts.
These quantized releases are published by Open4bits, the dedicated quantization project under ArkAiLabs, and are compatible with local inference tools such as llama.cpp, Ollama, and LM Studio on Windows, Linux, and macOS.
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.
The Llama-Nexora Family
This model is part of the Llama-Nexora family — a dedicated branch of Nexora models under ArkAiLabs, built on the Meta Llama architecture and focused on creative, efficient, and practical open AI systems.
| Model | Type | Link |
|---|---|---|
| llama-nexora-vector-v0.1 | Original (Full Precision) | ArkAiLab-Adl/llama-nexora-vector-v0.1 |
| llama-nexora-vector-v0.1-GGUF | GGUF (Windows / Linux / macOS) | (this repo) |
| llama-nexora-vector-v0.1-mlx-4Bit | MLX 4-Bit (Apple Silicon) | Open4bits/llama-nexora-vector-v0.1-mlx-4Bit |
Available Quantizations
All quantized files are available in this repository. Select the file that best matches your hardware and performance requirements.
2-bit
| Quantization | File Size | Description |
|---|---|---|
| Q2_K | 581 MB | Smallest size, lowest quality. Use only if very limited on RAM/VRAM. |
4-bit
| Quantization | File Size | Description |
|---|---|---|
| Q4_K_S | 776 MB | Small 4-bit quantization. Good balance of size and quality. |
| Q4_0 | 771 MB | Legacy 4-bit format. Widely compatible. |
| Q4_K_M | 808 MB | Medium 4-bit quantization. Recommended for most users. |
5-bit
| Quantization | File Size | Description |
|---|---|---|
| Q5_K_S | 893 MB | Small 5-bit quantization. Better quality than Q4 with modest size increase. |
| Q5_K_M | 912 MB | Medium 5-bit quantization. Excellent quality-to-size ratio. |
6-bit
| Quantization | File Size | Description |
|---|---|---|
| Q6_K | 1.02 GB | High quality, close to full precision. Recommended if you have the RAM. |
8-bit
| Quantization | File Size | Description |
|---|---|---|
| Q8_0 | 1.32 GB | Near full-precision quality. Best quality GGUF option available. |
Which Quant Should I Use?
| Your Situation | Recommended Quant |
|---|---|
| Very limited RAM (< 2GB free) | Q2_K |
| General use / most users | Q4_K_M |
| Want better quality, have the space | Q5_K_M or Q6_K |
| Maximum quality, no size concern | Q8_0 |
| Legacy tooling / broad compatibility | Q4_0 |
Tip: For most users, Q4_K_M offers the best balance between model size and output quality.
Model Details
| Property | Details |
|---|---|
| Model Name | llama-nexora-vector-v0.1-GGUF |
| Model Family | Llama-Nexora |
| Model Type | Text-to-SVG (Causal Language Model) |
| Original Base Model | unsloth/Llama-3.2-1B-Instruct |
| Original Full Model | ArkAiLab-Adl/llama-nexora-vector-v0.1 |
| Quantized By | Open4bits |
| Output Format | SVG |
| Release Status | Beta |
| License | Llama 3.2 Community License |
Requirements
GGUF models can be run on Windows, Linux, and macOS (including Apple Silicon and Intel) using any of the following tools:
- llama.cpp — CLI-based inference
- Ollama — Easy local model runner
- LM Studio — GUI-based local inference app
- Jan — Open-source ChatGPT alternative for local use
Installation & Usage
llama.cpp
# Clone and build llama.cpp
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make
# Download the model (example: Q4_K_M)
huggingface-cli download Open4bits/llama-nexora-vector-v0.1-GGUF \
llama-nexora-vector-v0.1.Q4_K_M.gguf \
--local-dir ./models
# Run inference
./llama-cli -m ./models/llama-nexora-vector-v0.1.Q4_K_M.gguf \
-p "Generate an SVG of a simple red circle." \
-n 512
Ollama
# Create a Modelfile
echo 'FROM ./llama-nexora-vector-v0.1.Q4_K_M.gguf' > Modelfile
# Create the model
ollama create llama-nexora-vector -f Modelfile
# Run it
ollama run llama-nexora-vector "Generate an SVG of a simple red circle."
LM Studio
- Open LM Studio and go to the Search tab.
- Search for
Open4bits/llama-nexora-vector-v0.1-GGUF. - Select your preferred quantization and download.
- Load the model and start prompting.
Capabilities
llama-nexora-vector-v0.1-GGUF 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 icons, shapes, logos, and simple illustrations
- Supporting rapid prototyping and concept design
- Producing lightweight scalable vector outputs
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 — dataset size affects output consistency across diverse prompts
- 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
- Quantization trade-off — lower-bit quants (Q2, Q4) may show more quality degradation versus the full-precision model
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 on local hardware
❌ 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
Usage Recommendations
To get the best results from this model:
- Keep prompts simple and specific — avoid multi-scene or highly complex compositions
- Validate all SVG outputs before rendering or integrating into any pipeline
- Post-process outputs to correct syntax or structural issues
- Use iterative prompting — refining prompts across multiple turns often yields better results
- Expect imperfections — this is a beta model; treat outputs as drafts, not finals
- Human review is recommended for all generated content
- Choose the right quant — higher-bit quants yield better output quality if your hardware allows
Risks & Considerations
Developers integrating this model 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
- Outputs may sometimes be invalid, incomplete, or require manual correction
- Quality degradation versus full-precision model, especially at lower bit widths
Recommendation: Implement downstream validation layers and SVG syntax checking before any rendering or integration. Human review is recommended for all generated content.
Community & Support
Join the community for updates, feedback, and discussion. Community feedback, testing, and contributions are welcome — this project will continue evolving through open research and real-world experimentation.
License
This model is released under the Llama 3.2 Community License.
Use of this model is governed by the Llama 3.2 Community License Agreement. Please review the license terms before use, modification, or distribution.
Acknowledgements
This quantized release is based on llama-nexora-vector-v0.1 by ArkAiLabs, which itself is built upon Llama 3.2 1B Instruct by Meta. Quantization was performed by Open4bits using the llama.cpp GGUF quantization toolchain. We thank the open-source AI community for their continued contributions that make projects like this possible.
About Open4bits
Open4bits is the official quantization project under ArkAiLabs, dedicated to publishing efficient, accessible quantized versions of Nexora and Llama-Nexora models across multiple formats (GGUF, MLX) for local inference on a wide range of hardware.
About Nexora & Llama-Nexora
Nexora is an experimental AI initiative under ArkAiLabs, focused on building lightweight, practical, and creative AI systems for real-world applications.
The Llama-Nexora family is a dedicated branch within Nexora, built on the Meta Llama architecture — focused on creative, efficient, and practical open AI systems that are accessible to the broader community.