On Fri, Apr 10, 2026 at 04:56:09PM +0200, Ali Farzanrad wrote:
>
> On 4/10/26 11:51 AM, Mike Larkin wrote:
> > On Fri, Apr 10, 2026 at 09:32:26AM +0000, Ali Farzanrad wrote: >> Hello
> > all, >> >> In my Minisforum UM790 Pro Mini PC, after running
> llama-cli it >> stuck at loading stage and my system gets very slow (it
> takes 4-5 >> seconds until I see the character after I press a key in
> xterm). >> However after disabling the amdgpu in /etc/bsd.re-config file I
> >> have no such problem. >> >> How can I debug the problem? >> > > what
> model are you using and what quantization? also please paste > the
> llama-cli/llama-server output. I've tested with these 2 models:
> Meta-Llama-3-8B-Instruct-Q4_K_M.gguf 4.6G
> Meta-Llama-3.1-8B-Instruct-Q5_K_M.gguf 5.3GI don't know how to verify that
Those seem to work fine here and get about 15 tok/s when asking it some
random question about a random event. You can see the build version
I was using below also. For the record, here's my cpu/gpu:
cpu0: AMD Ryzen 7 PRO 8840HS w/ Radeon 780M Graphics
▄ ▄▄
██ ██
██ ██ ▀▀█▄ ███▄███▄ ▀▀█▄ ▄████ ████▄ ████▄
██ ██ ▄█▀██ ██ ██ ██ ▄█▀██ ██ ██ ██ ██ ██
██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀
██ ██
▀▀ ▀▀
build : b8703-5c4aae66e
model : bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M
modalities : text
available commands:
/exit or Ctrl+C stop or exit
/regen regenerate the last response
/clear clear the chat history
/read <file> add a text file
/glob <pattern> add text files using globbing pattern
> tell me about the emu war
The Emu War was a unique and intriguing episode in Australian history that took
place in 1932. Here's what happened:
**Background:**
After World War I, many Australian soldiers returned to their homeland, but the
country was also experiencing the effects of the Great Depression. The economy
was in shambles, and jobs were scarce
...
...
[ Prompt: 121.3 t/s | Generation: 14.8 t/s ]
