Hello Everyone,
I have been trying to get Agent-Zero up and running on my configuration at home, but I am running into what I think is a timing issue with the LLMS. I am trying to get Agent-Zero working with LLMSTER (The headless version of LLM Studio). Here is my setup that I have tried but it can never respond just gives me an error. I will put a picture of the error here.
Setup (Running everything on a UnRaid ProServer 7.2.4 (With 256GB RAM, 8GB Nvidia card , with Nvidia Plugin):
Virtual Machine - LLMSTER Server
- 200GB RAM
- Virtual card
- 1 TB Hard Drive
- Ubuntu 24.04.4 Server
- What I have tried as the Chat and Utility models
- Qwen3.5 122B A10B [Q4_K_M] (75.12 GB)
- mistral-small-24b-instruct-2501
- Gemma-4-31B-it
- Gemma-4-26B-A4B-it
- I have also tried context 32768 to 16384 (For each one and matching Agent-Zero settings)
Docker Container
Agent-Zero
I have made sure that Agent-Zero can talk to the LLMSTER VM.
Any ideas or recommendations on which models to use? I wish I had a big NVIDIA card but I only have this server. with lots of ram. lol
Thanks in advance