Instructions to use Open4bits/llama-nexora-vector-v0.1-mlx-4Bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Open4bits/llama-nexora-vector-v0.1-mlx-4Bit 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("Open4bits/llama-nexora-vector-v0.1-mlx-4Bit")

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
LM Studio

How to use Open4bits/llama-nexora-vector-v0.1-mlx-4Bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Open4bits/llama-nexora-vector-v0.1-mlx-4Bit 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 "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit"

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 Open4bits/llama-nexora-vector-v0.1-mlx-4Bit

Run Hermes

hermes

OpenClaw new

How to use Open4bits/llama-nexora-vector-v0.1-mlx-4Bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit"

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 "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit" \
  --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"

MLX LM

How to use Open4bits/llama-nexora-vector-v0.1-mlx-4Bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "Open4bits/llama-nexora-vector-v0.1-mlx-4Bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

llama-nexora-vector-v0.1-mlx-4Bit / README.md

fmasterpro27

Update README.md

19a592b verified 2 months ago

preview code

Raw

History Blame Contribute Delete

9.81 kB

	---
	base_model:
	- ArkAiLab-Adl/llama-nexora-vector-v0.1
	license: llama3.2
	language:
	- en
	pipeline_tag: text-generation
	tags:
	- nexora
	- llama-nexora
	- vector
	- chat
	- llama-3
	- mlx
	- open4bits
	---
	<p align="center">
	<img src="https://huggingface.co/ArkAiLab-Adl/llama-nexora-vector-v0.1/resolve/main/assets/llama-nexora-vector.jpg" alt="llama-nexora-vector-mlx-4bit"/>
	</p>

	# Llama-Nexora-Vector-v0.1 — MLX 4-Bit

	<p align="center">
	<img src="https://img.shields.io/badge/status-beta-orange" alt="Status: Beta"/>
	<img src="https://img.shields.io/badge/license-Llama%203.2%20Community-blue" alt="License: Llama 3.2 Community"/>
	<img src="https://img.shields.io/badge/base_model-Llama--3.2--1B-blueviolet" alt="Base Model: Llama 3.2 1B"/>
	<img src="https://img.shields.io/badge/output-SVG-green" alt="Output: SVG"/>
	<img src="https://img.shields.io/badge/family-Llama--Nexora-red" alt="Family: Llama-Nexora"/>
	<img src="https://img.shields.io/badge/format-MLX%204--Bit-cyan" alt="Format: MLX 4-Bit"/>
	<img src="https://img.shields.io/badge/size-713MB-lightgrey" alt="Size: 713MB"/>
	</p>

	> This is the official MLX 4-bit quantized release of [llama-nexora-vector-v0.1](https://huggingface.co/ArkAiLab-Adl/llama-nexora-vector-v0.1), published by [Open4bits](https://huggingface.co/Open4bits) — the official quantization project under ArkAiLabs. This version is optimized for efficient inference on Apple Silicon (M1/M2/M3/M4) using the MLX framework. It is a beta release intended for research, prototyping, and early-stage development workflows only.

	---

	## Table of Contents

	- [Overview](#overview)
	- [The Llama-Nexora Family](#the-llama-nexora-family)
	- [Quantization Details](#quantization-details)
	- [Model Details](#model-details)
	- [Requirements](#requirements)
	- [Capabilities](#capabilities)
	- [Limitations](#limitations)
	- [Intended Use](#intended-use)
	- [Usage Recommendations](#usage-recommendations)
	- [Risks & Considerations](#risks--considerations)
	- [Community & Support](#community--support)
	- [License](#license)
	- [Acknowledgements](#acknowledgements)

	---

	## Overview

	llama-nexora-vector-v0.1-mlx-4Bit is the official MLX 4-bit quantized version of [llama-nexora-vector-v0.1](https://huggingface.co/ArkAiLab-Adl/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.

	This quantized release is published by [Open4bits](https://huggingface.co/Open4bits), the dedicated quantization project under ArkAiLabs, and is designed specifically for optimized local inference on Apple Silicon hardware via the [MLX](https://github.com/ml-explore/mlx) framework. The total model size is 713MB.

	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](https://huggingface.co/ArkAiLab-Adl/llama-nexora-vector-v0.1) \|
	\| llama-nexora-vector-v0.1-mlx-4Bit \| MLX 4-Bit (Apple Silicon) \| (this repo) \|

	> For the GGUF quantized version compatible with llama.cpp, Ollama, and LM Studio, visit [Open4bits](https://huggingface.co/Open4bits).

	---

	## Quantization Details

	\| Property \| Details \|
	\|---\|---\|
	\| Quantization Format \| MLX 4-Bit \|
	\| Quantized By \| [Open4bits](https://huggingface.co/Open4bits) (official ArkAiLabs quantization project) \|
	\| Original Model \| [ArkAiLab-Adl/llama-nexora-vector-v0.1](https://huggingface.co/ArkAiLab-Adl/llama-nexora-vector-v0.1) \|
	\| Model Size \| 713MB \|
	\| Target Platform \| Apple Silicon (M1/M2/M3/M4) \|
	\| Framework \| [MLX](https://github.com/ml-explore/mlx) \|

	---

	## Model Details

	\| Property \| Details \|
	\|---\|---\|
	\| Model Name \| llama-nexora-vector-v0.1-mlx-4Bit \|
	\| Model Family \| Llama-Nexora \|
	\| Model Type \| Text-to-SVG (Causal Language Model) \|
	\| Original Base Model \| [unsloth/Llama-3.2-1B-Instruct](https://huggingface.co/unsloth/Llama-3.2-1B-Instruct) \|
	\| Original Full Model \| [ArkAiLab-Adl/llama-nexora-vector-v0.1](https://huggingface.co/ArkAiLab-Adl/llama-nexora-vector-v0.1) \|
	\| Output Format \| SVG \|
	\| Release Status \| Beta \|
	\| License \| Llama 3.2 Community License \|

	---

	## Requirements

	- Hardware: Apple Silicon Mac (M1, M2, M3, or M4)
	- OS: macOS 13.3 or later
	- Framework: [MLX](https://github.com/ml-explore/mlx) and [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms)

	---

	## Capabilities

	llama-nexora-vector-v0.1-mlx-4Bit 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 — 4-bit quantization may introduce minor degradation in output quality compared to the full-precision model

	---

	## Intended Use

	### ✅ Supported Use Cases

	- Academic and applied research in text-to-vector generation
	- Experimental AI-assisted design systems on Apple Silicon
	- Educational exploration of structured output generation
	- Lightweight SVG prototyping and ideation on local Mac 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
	- Non-Apple Silicon hardware (use the GGUF version instead)

	---

	## Usage Recommendations

	To get the best results from this model:

	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
	6. Human review is recommended for all generated content

	---

	## 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
	- Minor quality degradation versus the full-precision model due to 4-bit quantization

	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.

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

	---

	## 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](https://www.llama.com/llama3_2/license/). Please review the license terms before use, modification, or distribution.

	---

	## Acknowledgements

	This quantized release is based on [llama-nexora-vector-v0.1](https://huggingface.co/ArkAiLab-Adl/llama-nexora-vector-v0.1) by ArkAiLabs, which itself is built upon [Llama 3.2 1B Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) by Meta. Quantization was performed by [Open4bits](https://huggingface.co/Open4bits) using the [MLX](https://github.com/ml-explore/mlx) framework. We thank the open-source AI community for their continued contributions that make projects like this possible.

	---

	## About Open4bits

	[Open4bits](https://huggingface.co/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.