Browse categories
Explore
Fiverr Pro
English
$
USD
I will setup deepseek harness and local ai agent with llama cpp
Lucas L.
About this gig
Turn your GPU into a private AI agent no per-token fees.
>
DeepSeek Harness + llama.cpp is the most powerful open-source local AI stack. Harness gives you an agent framework with plugin support; llama.cpp gives you inference that's 2-5x faster than Ollama.
>
I'll set it all up for you in one session.
>
What you get:
- DeepSeek Harness (dsh) installed and configured
- llama.cpp compiled from source with GPU acceleration
- llama-server running as a production REST API
- The right model sized for your VRAM
>
Stack: DeepSeek Harness + llama-server + GGUF models (Qwen, Gemma, DeepSeek, Llama, Mistral...)
>
What I need: Your OS, GPU specs, and which model you want.
Get to know Lucas L.
Lucas L.
Remote IT Support Windows Printers Virtual Machines Local AI Setup
- FromArgentina
- Member sinceNov 2025
- Avg. response time1 hour
- Last delivery1 month
Languages
English, Spanish
I'm a remote IT support specialist with 10+ years of hands-on experience in Windows environments, networking, and technical support.
I help individuals and small businesses with:
• Virtual machine setup (VMware, VirtualBox, Hyper-V)
• Local AI installation (Ollama, LM Studio, DeepSeek)
• Printer installation & troubleshooting (USB, WiFi, LAN)
• Windows repairs, driver fixes, and PC optimization
Every session is remote, fast, and frustration-free. If I can't fix it, you pay nothing.
Message me before ordering if you're unsure which service fits your needs — I'm here to help.
My Portfolio
FAQ
Will this work on my hardware?
Share your GPU (NVIDIA/AMD/Apple Silicon) or CPU-only setup before ordering. I'll confirm
Which model should I run?
I'll recommend one based on your VRAM. Larger models = better quality, more memory needed
Ollama is easier. Why llama.cpp?
llama.cpp is 2-5x faster and gives you full control. Harness works with both, but the llama.cpp backend is the performance winner.
