Please need help to config OpenClaw Pre-Installed model: Beelink SER10 MAX AMD Pro Ryzen™ AI 9 HX 470 (OpenClaw & Local LLM Pre-installed), i follow the manual, on /home but i can’t make work llama.cpp with openclaw (basic, without any integration Telegram, whasta app etc . yet ), llama.cpp works fine on port 8080 i try several options on external forums to beelink with not positive results (infinite answer) on chat openclaw . set time to response to agent more high. embedded run agent end or embedded run failover decision disable firewall ubuntu ( maybe not convenient). update openclaw to version 2026.4.22 not work (Bootstrap pending), any as infinite time to answer. reconfigure openclaw “ openclaw onboard –install-daemon” and “ openclaw doctor –fix” dont work. Please, need a some help to meet the goal, thanks in advance.

Configure SER10 OpenClaw Pre-Installed model

juanchito51

Please need help to config OpenClaw Pre-Installed model: Beelink SER10 MAX AMD Pro Ryzen™ AI 9 HX 470 (OpenClaw & Local LLM Pre-installed), i follow the manual, on /home but i can’t make work llama.cpp with openclaw (basic, without any integration Telegram, whasta app etc . yet ),
llama.cpp works fine on port 8080
i try several options on external forums to beelink with not positive results (infinite answer) on chat openclaw .

set time to response to agent more high. embedded run agent end or embedded run failover decision
disable firewall ubuntu ( maybe not convenient).
update openclaw to version 2026.4.22 not work (Bootstrap pending), any as infinite time to answer.
reconfigure openclaw “ openclaw onboard –install-daemon” and “ openclaw doctor –fix” dont work.

Please, need a some help to meet the goal, thanks in advance.

Beelink CS-June

juanchito51

Hello there,

Please select your AI model provider: choose vLLM for a locally hosted model, or select a cloud provider by name.

Please set the vLLM base URL port to 8080 (llama.cpp default) or 11434 (Ollama default). The API key field can be any placeholder value. For the vLLM model field, enter the ID of the model you have loaded

Have a nice day~