Door categorieën bladeren
Ontdekken
Fiverr Pro
Nederlands
$
USD
I will run llms on your mac and self host open source apps
Amila Thushara
Over deze dienst
Want to run AI on your own hardware, private, secure, and with no monthly API bills?
You're in the right place!
I set up self-hosted LLMs and open-source AI tools on your Mac Studio, Mac cluster, GPU server, or on-prem machines, fully private and ready to use.
What I Offer:
- Local LLM setup (Llama, Qwen, DeepSeek, Mistral, Gemma)
- Mac Studio & Mac Mini LLM setup with MLX
- Mac cluster setup for large models (exo, MLX)
- GPU server & on-prem LLM deployment
- Open WebUI private ChatGPT setup
- n8n AI automation self-hosting
- Private RAG chatbot on your own documents
- Dify, Flowise & AnythingLLM setup
- Secure remote access & user management
- Model quantization & performance tuning
Tech Stack:
LLM Runtimes: Ollama || llama.cpp || MLX || vLLM || LM Studio || exo
Open-Source Apps: Open WebUI || n8n || Dify || Flowise || AnythingLLM || LocalAI
Hardware: Mac Studio || Mac Mini || NVIDIA GPU Servers || On-Prem
Tools: Docker || Linux || Tailscale || Nginx
Why Choose Me?
- Fast delivery
- Free consultation
- 100% private, data stays on your hardware
- Full documentation
- Post-setup support
Let's run your own AI today!
Maak kennis met Amila Thushara
Amila Thushara
Senior Software Engineer POS, ERP and AI Powered Web Mobile Solutions
- Afkomstig uitSri Lanka
- Lid sindsjun 2020
- Gem. reactietijd2 uur
Talen
Engels
Senior Software Engineer building scalable POS, ERP, MVP, web & mobile apps with a focus on performance, security & maintainability. Skilled in microservices, API development, database optimization, system integrations & AI-powered features (chatbots, automation, LLM integrations). From idea validation to production launch & long-term support, I handle every project with ownership and a results-driven mindset. Let's build something reliable together — message me to discuss your project.
Andere AI-development diensten die ik aanbied
Veelgestelde vragen
What hardware do I need to run a local LLM?
Small models (7B to 8B) run well on a Mac with 16GB+ RAM or a GPU with 8GB+ VRAM. Larger models (70B+) need a Mac Studio with 64GB+ or a high-VRAM GPU server. I'll recommend the right model for your hardware.
Can you set up LLMs on a Mac Studio or Mac Mini?
Yes. Apple Silicon Macs are excellent for local AI thanks to unified memory. I use MLX, Ollama, and llama.cpp to get the best speed on M-series chips.
What is a Mac cluster, and do I need one?
A Mac cluster connects multiple Macs to run models too large for one machine, like 400B+ parameter models. Using tools like exo and MLX, I combine their memory so they work as one AI system.
What is Open WebUI?
Open WebUI is a private, ChatGPT-style interface for your local LLMs. Your team can chat, upload documents, and manage conversations, all on your own hardware with user accounts and permissions.
Can you self-host n8n for AI automation?
Yes. I set up n8n on your server and connect it to your local LLM, so you can automate emails, documents, and workflows with AI without paying per-task fees or sharing data.
Is my data really private?
Yes. Everything runs on your own machines, so prompts, documents, and data never leave your network or go to OpenAI, Google, or any third party.
Can I access my local AI remotely?
Yes. I can set up secure remote access with Tailscale, a VPN, or Nginx with SSL, so you and your team can use it from anywhere safely.
Will a local LLM be as good as ChatGPT?
Modern open-source models like Llama, Qwen, and DeepSeek perform very well for most business tasks. I'll help you choose the best model for your needs and hardware.
How do you access my machine for setup?
Through secure remote access you approve, such as SSH, AnyDesk, or a screen-sharing session. I never need your personal passwords, and you can revoke access once we're done.
Not sure which package fits your setup?
Message me before ordering with your hardware details. I'll recommend the right package or send a custom offer.
