View source for LLM/Ryzen AI 7 350/2026-07-09

= Install lemonade from PPA =

<syntaxhighlight lang="bash">
sudo add-apt-repository ppa:lemonade-team/stable
sudo apt update
sudo apt install lemonade-server
</syntaxhighlight>

== Run lemonade server ==

<syntaxhighlight lang="bash">
sudo systemctl start lemond
sudo systemctl enable lemond
</syntaxhighlight>

== Install Backends ==

Open http://localhost:13305, then ...

=== Install llama.cpp cuda ===

* Click backends at left sidebar.
* Find Llama.cpp GPU
* Find child item cuda
* Click download

<syntaxhighlight lang="bash">
sudo vim /etc/gdm3/custom.conf
</syntaxhighlight>

<pre>
[daemon]
# 強制 GDM 的 Wayland 渲染走 AMD 的基底
Environment=VK_LOADER_DRIVERS_SELECT=*radeon*
Environment=MESA_VK_DEVICE_SELECT=pci-0000_66_00_0
</pre>

<syntaxhighlight lang="bash">
sudo vim /etc/default/grub
</syntaxhighlight>

<pre>
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nvidia-drm.modeset=0"
</pre>

<syntaxhighlight lang="bash">
sudo update-grub
</syntaxhighlight>

=== Install Fastflow ===

* Click backends at left sidebar.
* Find FastFlowLM NPU
* Find child item npu
* Click download
* Install XDNA driver

<syntaxhighlight lang="bash">
sudo apt install amdxdna-dkms
sudo reboot
</syntaxhighlight>

* Get FastFlowLM release from https://github.com/FastFlowLM/FastFlowLM/releases/
* Download the latest db for Ubuntu 26.04

<syntaxhighlight lang="bash">
cd 下載
sudo apt install ./fastflowlm_0.9.44_ubuntu26.04_amd64.deb
flm --version
flm validate
</syntaxhighlight>

<pre>
Memlock limit is too low (8MB). Please ...
</pre>

=== Fix memory limit problem ===

<syntaxhighlight lang="bash">
sudo nano /etc/security/limits.conf
</syntaxhighlight>

Add before `# End of file`

<pre>
* soft memlock unlimited
* hard memlock unlimited
</pre>

<syntaxhighlight lang="bash">
sudo reboot
</syntaxhighlight>

<syntaxhighlight lang="bash">
ulimit -l
flm validate
</syntaxhighlight>

=== Enable NPU and set default models ===

<syntaxhighlight lang="bash">
sudo systemctl edit lemond
</syntaxhighlight>

<pre>
[Service]
# Enable NPU
Environment="LEMONADE_ENABLE_NPU=true"
# Prevent from model unloaded
Environment="LEMONADE_MODEL_TTL=0"
# Default models
ExecStartPost=/usr/bin/lemonade load model --ctx-size 65535 Gemma-4-12B-it-GGUF --backend cuda
ExecStartPost=/usr/bin/lemonade load model gpt-oss-sg-20b-FLM --backend npu
</pre>

<syntaxhighlight lang="bash">
sudo systemctl daemon-reload
sudo systemctl restart lemond
</syntaxhighlight>

=== Let iGPU can use large memory size ===

Default iGPU memory limit is 536.87MB in Ubuntu 26.04

<syntaxhighlight lang="bash">
sudo nano /etc/modprobe.d/amdgpu.conf
</syntaxhighlight>

<pre>
options amdgpu ttm_pages_limit=25600000
options amdgpu page_pool_size=25600000
</pre>

<syntaxhighlight lang="bash">
sudo update-initramfs -u
sudo reboot
sudo dmesg | grep -iE "amdgpu.*memory"
</syntaxhighlight>

<pre>
[    5.304706] amdgpu 0000:66:00.0: Trusted Memory Zone (TMZ) feature disabled as experimental (default)
[    5.304936] amdgpu 0000:66:00.0:  512M of VRAM memory ready
[    5.304939] amdgpu 0000:66:00.0:  46677M of GTT memory ready.
</pre>

sudo apt install vulkan-tools
sudo vim /var/lib/lemonade/.cache/lemonade/config.json

== Optimize Lemonade config ==

Find Cache dir

<syntaxhighlight lang="bash">
sudo journalctl -u lemond -f -n 100
</syntaxhighlight>

<pre>
 7月 09 17:37:22 agentlab-zen5 lemond[411649]: 2026-07-09 17:37:22.869 [Info] (main) Starting Lemonade Server...
 7月 09 17:37:22 agentlab-zen5 lemond[411649]: 2026-07-09 17:37:22.869 [Info] (main)   Version: 10.9.0
 7月 09 17:37:22 agentlab-zen5 lemond[411649]: 2026-07-09 17:37:22.869 [Info] (main)   Cache dir: /var/lib/lemonade/.cache/lemonade
 7月 09 17:37:22 agentlab-zen5 lemond[411649]: 2026-07-09 17:37:22.869 [Info] (main)   Port: 13305
 7月 09 17:37:22 agentlab-zen5 lemond[411649]: 2026-07-09 17:37:22.869 [Info] (main)   Host: localhost
 7月 09 17:37:22 agentlab-zen5 lemond[411649]: 2026-07-09 17:37:22.869 [Info] (main)   Log level: info
 7月 09 17:37:22 agentlab-zen5 lemond[411649]: 2026-07-09 17:37:22.869 [Info] (main)   Telemetry: disabled
</pre>

Edit config.json

<syntaxhighlight lang="bash">
sudo vim /var/lib/lemonade/.cache/lemonade/config.json
</syntaxhighlight>

* host: 0.0.0.0
* max_loaded_models: -1

<syntaxhighlight lang="bash">
sudo systemctl restart lemond
</syntaxhighlight>

Now Lemonade can load multiple models.

== Debug ==

<syntaxhighlight lang="bash">
sudo journalctl -u lemond.service -n 50 --no-pager
sudo journalctl -u lemond -f -n 100
</syntaxhighlight>