NodeBB

J’ai pas trouvé de bon process pour la purge, donc ma méthode :

# uptrack-uname -r
5.4.17-2136.349.3.1.el8uek.x86_64
#  uptrack-show
Installed updates:
None

Effective kernel version is 5.4.17-2136.349.3.1.el8uek

# du -sh /var/cache/uptrack/Linux/x86_64/*
449M    /var/cache/uptrack/Linux/x86_64/5.4.17-2102.201.3.el8uek.x86_64
287M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.316.7.el8uek.x86_64
211M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.324.5.3.el8uek.x86_64
214M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.328.3.el8uek.x86_64
456M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.331.7.el8uek.x86_64
132M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.335.4.1.el8uek.x86_64
270M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.336.5.1.el8uek.x86_64
604M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.337.5.1.el8uek.x86_64
126M    /var/cache/uptrack/Linux/x86_64/5.4.17-2136.349.3.1.el8uek.x86_64
#  yum remove uptrack
Dependencies resolved.
=============================================================================================================================================================================================================================================
 Package                                                          Architecture                                     Version                                                  Repository                                                  Size
=============================================================================================================================================================================================================================================
Removing:
 uptrack                                                          noarch                                           1.2.80-0.el8                                             @ksplice-uptrack                                           626 k
Removing unused dependencies:
 perl-IPC-System-Simple                                           noarch                                           1.25-17.el8                                              @ol8_appstream                                              69 k
 perl-autodie                                                     noarch                                           2.29-396.el8                                             @ol8_appstream                                             211 k
 python3-pycurl                                                   x86_64                                           7.43.0.2-4.el8                                           @ol8_appstream                                             767 k

Transaction Summary
=============================================================================================================================================================================================================================================
Remove  4 Packages

Freed space: 1.6 M
Is this ok [y/N]: y
Running transaction check
Transaction check succeeded.
Running transaction test
Transaction test succeeded.
Running transaction
  Preparing        :                                                                                                                                                                                                                     1/1
  Running scriptlet: uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         1/1
  Running scriptlet: uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         1/4
  Erasing          : uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         1/4
warning: /etc/uptrack/uptrack.conf saved as /etc/uptrack/uptrack.conf.rpmsave

  Running scriptlet: uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         1/4
  Erasing          : perl-autodie-2.29-396.el8.noarch                                                                                                                                                                                    2/4
  Erasing          : perl-IPC-System-Simple-1.25-17.el8.noarch                                                                                                                                                                           3/4
  Erasing          : python3-pycurl-7.43.0.2-4.el8.x86_64                                                                                                                                                                                4/4
  Running scriptlet: python3-pycurl-7.43.0.2-4.el8.x86_64                                                                                                                                                                                4/4
  Verifying        : perl-IPC-System-Simple-1.25-17.el8.noarch                                                                                                                                                                           1/4
  Verifying        : perl-autodie-2.29-396.el8.noarch                                                                                                                                                                                    2/4
  Verifying        : python3-pycurl-7.43.0.2-4.el8.x86_64                                                                                                                                                                                3/4
  Verifying        : uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         4/4

Removed:
  perl-IPC-System-Simple-1.25-17.el8.noarch                         perl-autodie-2.29-396.el8.noarch                         python3-pycurl-7.43.0.2-4.el8.x86_64                         uptrack-1.2.80-0.el8.noarch

Complete!
#  yum install uptrack
Extra Packages for Enterprise Linux 8 - x86_64                                                                                                                                                               4.1 MB/s |  14 MB     00:03
Ksplice Uptrack for Oracle Linux                                                                                                                                                                             4.6 kB/s | 5.0 kB     00:01
packages-microsoft-com-prod                                                                                                                                                                                   43 MB/s |  20 MB     00:00
Oracle Linux 8 BaseOS Latest (x86_64)                                                                                                                                                                         51 MB/s | 126 MB     00:02
Oracle Linux 8 Application Stream (x86_64)                                                                                                                                                                    48 MB/s |  77 MB     00:01
Latest Unbreakable Enterprise Kernel Release 6 for Oracle Linux 8 (x86_64)                                                                                                                                    49 MB/s | 136 MB     00:02
Dependencies resolved.
=============================================================================================================================================================================================================================================
 Package                                                          Architecture                                     Version                                                   Repository                                                 Size
=============================================================================================================================================================================================================================================
Installing:
 uptrack                                                          noarch                                           1.2.80-0.el8                                              ksplice-uptrack                                           160 k
Installing dependencies:
 perl-IPC-System-Simple                                           noarch                                           1.25-17.el8                                               ol8_appstream                                              43 k
 perl-autodie                                                     noarch                                           2.29-396.el8                                              ol8_appstream                                              98 k
 python3-pycurl                                                   x86_64                                           7.43.0.2-4.el8                                            ol8_appstream                                             227 k

Transaction Summary
=============================================================================================================================================================================================================================================
Install  4 Packages

Total download size: 528 k
Installed size: 1.6 M
Is this ok [y/N]: y
Downloading Packages:
(1/4): perl-autodie-2.29-396.el8.noarch.rpm                                                                                                                                                                  1.0 MB/s |  98 kB     00:00
(2/4): perl-IPC-System-Simple-1.25-17.el8.noarch.rpm                                                                                                                                                         433 kB/s |  43 kB     00:00
(3/4): python3-pycurl-7.43.0.2-4.el8.x86_64.rpm                                                                                                                                                               10 MB/s | 227 kB     00:00
(4/4): uptrack-1.2.80-0.el8.noarch.rpm                                                                                                                                                                       199 kB/s | 160 kB     00:00
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Total                                                                                                                                                                                                        655 kB/s | 528 kB     00:00
Running transaction check
Transaction check succeeded.
Running transaction test
Transaction test succeeded.
Running transaction
  Preparing        :                                                                                                                                                                                                                     1/1
  Installing       : python3-pycurl-7.43.0.2-4.el8.x86_64                                                                                                                                                                                1/4
  Installing       : perl-IPC-System-Simple-1.25-17.el8.noarch                                                                                                                                                                           2/4
  Installing       : perl-autodie-2.29-396.el8.noarch                                                                                                                                                                                    3/4
  Installing       : uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         4/4
  Running scriptlet: uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         4/4
There are no existing modules on disk that need basename migration.

  Verifying        : uptrack-1.2.80-0.el8.noarch                                                                                                                                                                                         1/4
  Verifying        : perl-IPC-System-Simple-1.25-17.el8.noarch                                                                                                                                                                           2/4
  Verifying        : perl-autodie-2.29-396.el8.noarch                                                                                                                                                                                    3/4
  Verifying        : python3-pycurl-7.43.0.2-4.el8.x86_64                                                                                                                                                                                4/4

Installed:
  perl-IPC-System-Simple-1.25-17.el8.noarch                         perl-autodie-2.29-396.el8.noarch                         python3-pycurl-7.43.0.2-4.el8.x86_64                         uptrack-1.2.80-0.el8.noarch

Complete!
# du -sh /var/cache/uptrack/
0       /var/cache/uptrack/

A suivre.

Test des fichiers robots.txt :

https://www.cyber-neurones.org/robots.txt => OK.
https://lemmy.cyber-neurones.org/robots.txt => OK.
https://git.cyber-neurones.org/robots.txt => OK.
https://move.cyber-neurones.org/robots.txt => OK.
https://aventures.cyber-neurones.org/robots.txt => KO.
http://stats.cyber-neurones.org/robots.txt => OK ( vers www )
https://sat.cyber-neurones.org/robots.txt => OK ( vers www )
https://insta.cyber-neurones.org/robots.txt => OK .
https://webanalyse.cyber-neurones.org/robots.txt => OK ( vers www )
https://timeline.cyber-neurones.org/robots.txt => OK.
https://glance.cyber-neurones.org/robots.txt => KO.
https://grafana.cyber-neurones.org/robots.txt => OK.
https://healthchecks.cyber-neurones.org/robots.txt ( vers www )
http://healthchecks.cyber-neurones.org/robots.txt => KO.

Test des fichiers sitemap.xml :

http://www.cyber-neurones.org/sitemap.xml => OK ?
https://www.cyber-neurones.org/wp-sitemap.xml => OK.
https://lemmy.cyber-neurones.org/sitemap.xml => OK.
https://git.cyber-neurones.org/sitemap.xml => OK.
https://aventures.cyber-neurones.org/sitemap.xml => KO.
http://stats.cyber-neurones.org//sitemap.xml => KO.
https://sat.cyber-neurones.org/sitemap.xml => KO.
https://insta.cyber-neurones.org/sitemap.xml => KO.

Test : https://www.xml-sitemaps.com/validate-xml-sitemap.html

Augmentation du contexte :

  -m /models/Qwen3.6-27B-Q4_K_M.gguf --mmproj /models/mmproj-BF16.gguf --ctx-size 80000 

Update de la configuration : https://unsloth.ai/docs/models/qwen3.6#mtp-qwen3.6-27b

ExecStart=/usr/local/bin/llama-server \
  -m /models/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf --mmproj /models/mmproj-F32.gguf --ctx-size 140000 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00 \
  --host 0.0.0.0 --port 8080

Installation : https://hermes-ai.net/fr/docs/quickstart/

• Hermes : Hermes Agent v0.18.0 (2026.7.1) · upstream 179ca25a · local 2c5762f5 (+1 carried commit)
• OS : 22.04.5 LTS (Jammy Jellyfish)
• llama.cpp : 9687 (b4024af6c)
  • /models/Qwen3.6-27B-Q4_K_M.gguf ( https://huggingface.co/unsloth/Qwen3.6-27B-GGUF/blob/f31e38426bd3f5be50ca5bca4331e0f4b036b2ef/Qwen3.6-27B-Q4_K_M.gguf )
  • /models/mmproj-BF16.gguf
  • ctx-size 80000
• NVIDIA-SMI 595.71.05
• Driver Version: 595.71.05
• CUDA Version: 13.2
• PC : Laptop Tuxedo :
  • 11th Gen Intel Core i7-11800H
  • NVIDIA GeForce RTX 3060

$ inxi -C
CPU:
  Info: 8-core model: 11th Gen Intel Core i7-11800H bits: 64 type: MT MCP
    cache: L2: 10 MiB
  Speed (MHz): avg: 1987 min/max: 800/4600 cores: 1: 2700 2: 800 3: 2700
    4: 2700 5: 800 6: 2700 7: 2700 8: 2700 9: 2700 10: 2700 11: 800 12: 800
    13: 2700 14: 2700 15: 800 16: 800

$ llama-server --version
version: 9687 (b4024af6c)
built with GNU 11.4.0 for Linux x86_64

Doc : https://hermes-agent.nousresearch.com/docs/user-guide/sessions#export-sessions

# hermes sessions list
Preview                                            Last Active   Src    ID
───────────────────────────────────────────────────────────────────────────────────────────────
Il faut m’extraire toutes les randonnées du site   4m ago        cli    20260710_124146_504474
Il faut m’extraire toutes les randonnées du site   3h ago        cli    20260710_114254_75aa85
Merci de m'extraire toutes les randonnées du sit   4h ago        cli    20260710_094032_82cca6
Merci de m'extraire toutes les randonnées du sit   6h ago        cli    20260710_093835_ebdc3b
Merci de m'extraire toutes les randonnées du sit   6h ago        cli    20260710_092740_29d7a7
Merci de m'extraire toutes les randonnées du sit   6h ago        cli    20260710_091404_1f21cb
I want a list on all treks ( Randonnée in french   7h ago        cli    20260708_093212_f292fa
Give me the list in Markdown of all Trek on http   2d ago        cli    20260707_160320_a2dd13
Give me the list in Markdown of all Trek on http   3d ago        cli    20260707_144837_f45477
Give me the list in Markdown of all Trek on http   3d ago        cli    20260707_144412_8aa574
test                                               3d ago        cli    20260707_112602_4e4c33
# hermes sessions export md
Exported 11 sessions to md

L’export en Markdown ne fonctionne pas :

# hermes sessions export --format md --older-than 90 --dry-run
usage: hermes [-h] [--version] [-z PROMPT] [--usage-file PATH] [-m MODEL] [--provider PROVIDER]
              [-t TOOLSETS] [--resume SESSION] [--continue [SESSION_NAME]] [--worktree] [--accept-hooks]
              [--skills SKILLS] [--yolo] [--pass-session-id] [--ignore-user-config] [--ignore-rules]
              [--safe-mode] [--tui] [--cli] [--dev]
              {chat,model,moa,fallback,secrets,migrate,gateway,proxy,lsp,setup,postinstall,whatsapp,whatsapp-cloud,slack,send,login,logout,auth,status,cron,webhook,portal,kanban,project,hooks,doctor,security,dump,debug,backup,checkpoints,import,config,console,pairing,skills,bundles,plugins,curator,pets,journey,learning,memory-graph,memory,tools,computer-use,mcp,sessions,insights,claw,version,update,uninstall,acp,profile,completion,dashboard,serve,desktop,gui,logs,prompt-size}
              ...
hermes: error: unrecognized arguments: --format --older-than 90 --dry-run

Configuration :

Version de Hermes : Hermes Agent v0.18.2 (2026.7.7.2)

Dans /root/.hermes/config.yaml

model:
  default: local
  provider: custom
  base_url: http://127.0.0.1:8080/

Dans /etc/systemd/system/llama-server.service

ExecStart=/usr/local/bin/llama-server \
  -m /models/Qwen3.6-27B-Q3_K_M.gguf --mmproj /models/mmproj-BF16.gguf \
    --ctx-size 200000 \
    --temp 1.0 \
    --device CUDA0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00 \
  --host 0.0.0.0 --port 8080

Version Ubuntu :

# lsb_release -a
No LSB modules are available.
Distributor ID: Tuxedo
Description:    TUXEDO OS
Release:        24.04
Codename:       noble

Nouveau test :

# llama-bench -m /models/Qwen3.6-27B-Q3_K_M.gguf -dev CUDA0 -fa on --threads 12 -ngl 999 --cache-type-v q8_0
ggml_cuda_init: found 2 CUDA devices (Total VRAM: 21692 MiB):
  Device 0: NVIDIA GeForce RTX 5060 Ti, compute capability 12.0, VMM: yes, VRAM: 15888 MiB
  Device 1: NVIDIA GeForce RTX 3060 Laptop GPU, compute capability 8.6, VMM: yes, VRAM: 5803 MiB
| model                          |       size |     params | backend    | ngl | threads | type_v |  fa | dev          |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | ------: | -----: | --: | ------------ | --------------: | -------------------: |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       | 999 |      12 |   q8_0 |   1 | CUDA0        |           pp512 |        387.17 ± 2.25 |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       | 999 |      12 |   q8_0 |   1 | CUDA0        |           tg128 |         18.31 ± 0.01 |

# llama-bench -m /models/Qwen3.6-27B-Q3_K_M.gguf -fa on --threads 12 -ngl 999 --cache-type-v q8_0
ggml_cuda_init: found 2 CUDA devices (Total VRAM: 21692 MiB):
  Device 0: NVIDIA GeForce RTX 5060 Ti, compute capability 12.0, VMM: yes, VRAM: 15888 MiB
  Device 1: NVIDIA GeForce RTX 3060 Laptop GPU, compute capability 8.6, VMM: yes, VRAM: 5803 MiB
| model                          |       size |     params | backend    | ngl | threads | type_v |  fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | ------: | -----: | --: | --------------: | -------------------: |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       | 999 |      12 |   q8_0 |   1 |           pp512 |        367.28 ± 1.88 |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       | 999 |      12 |   q8_0 |   1 |           tg128 |         16.71 ± 0.03 |

Nouveau test avec /models/Qwen3.6-27B-Q3_K_M.gguf :

# llama-bench -m /models/Qwen3.6-27B-Q3_K_M.gguf -dev CUDA1
ggml_cuda_init: found 2 CUDA devices (Total VRAM: 21692 MiB):
  Device 0: NVIDIA GeForce RTX 5060 Ti, compute capability 12.0, VMM: yes, VRAM: 15888 MiB
  Device 1: NVIDIA GeForce RTX 3060 Laptop GPU, compute capability 8.6, VMM: yes, VRAM: 5803 MiB
| model                          |       size |     params | backend    | ngl | dev          |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | ------------ | --------------: | -------------------: |
llama_bench: error: failed to load model '/models/Qwen3.6-27B-Q3_K_M.gguf'

# llama-bench -m /models/Qwen3.6-27B-Q3_K_M.gguf 
ggml_cuda_init: found 2 CUDA devices (Total VRAM: 21692 MiB):
  Device 0: NVIDIA GeForce RTX 5060 Ti, compute capability 12.0, VMM: yes, VRAM: 15888 MiB
  Device 1: NVIDIA GeForce RTX 3060 Laptop GPU, compute capability 8.6, VMM: yes, VRAM: 5803 MiB
| model                          |       size |     params | backend    | ngl |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | --------------: | -------------------: |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       |  -1 |           pp512 |        707.38 ± 5.11 |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       |  -1 |           tg128 |         18.31 ± 0.01 |

build: c3d47e696 (10030)

# llama-bench -m /models/Qwen3.6-27B-Q3_K_M.gguf -dev CUDA0
ggml_cuda_init: found 2 CUDA devices (Total VRAM: 21692 MiB):
  Device 0: NVIDIA GeForce RTX 5060 Ti, compute capability 12.0, VMM: yes, VRAM: 15888 MiB
  Device 1: NVIDIA GeForce RTX 3060 Laptop GPU, compute capability 8.6, VMM: yes, VRAM: 5803 MiB
| model                          |       size |     params | backend    | ngl | dev          |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | ------------ | --------------: | -------------------: |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       |  -1 | CUDA0        |           pp512 |        841.06 ± 9.58 |
| qwen35 27B Q3_K - Medium       |  12.64 GiB |    26.90 B | CUDA       |  -1 | CUDA0        |           tg128 |         20.50 ± 0.02 |

build: c3d47e696 (10030)

Pendant le test sur les deux cartes j’ai lancé gpustat , j’ai les deux cartes qui tournent :

# gpustat 
tuxedo-17                              Fri Jul 17 17:14:51 2026  610.43.03
[0] NVIDIA GeForce RTX 3060 Laptop GPU | 58°C,  40 % |  4156 /  6144 MB | root(4148M)
[1] NVIDIA GeForce RTX 5060 Ti         | 62°C,  60 % |  8916 / 16311 MB | root(8908M)

Quand je suis sur CUDA0 :

 gpustat 
tuxedo-17                              Fri Jul 17 17:15:48 2026  610.43.03
[0] NVIDIA GeForce RTX 3060 Laptop GPU | 49°C,   0 % |     2 /  6144 MB |
[1] NVIDIA GeForce RTX 5060 Ti         | 68°C, 100 % | 12882 / 16311 MB | root(12874M)

La configuration finale :

Hermes: /root/.hermes/config.yaml

model:
  default: "llama-swap"
  provider: custom
  base_url: http://127.0.0.1:8080/v1
  api_key: "llama-swap"

Llama-swap : /etc/llama-swap/config.yaml

healthCheckTimeout: 500
logLevel: info
logTimeFormat: ""
logToStdout: "proxy"

# store: persistent storage for activity metrics state
# - optional, default: in-memory sqlite database capped by metricsMaxInMemory
# - path is a sqlite database file path
# - file-backed sqlite keeps activity logs across restarts
# store:
#   path: /path/to/file.sqlite

metricsMaxInMemory: 1000
captureBuffer: 15
ui:
  activity:
    session_id: ["X-Session-ID", "X-Litellm-Session-Id"]

performance:
  disabled: false
  every: 15s

startPort: 10001

sendLoadingState: true

includeAliasesInList: false

globalTTL: 0

unloadTimeout: 10

macros:
  "default_ctx": 40000
  "default_args": "--ctx-size ${default_ctx}"

models:
  "Best_Qwen3.6-27B-Q3_K_M.gguf_mmproj-BF16.gguf":
    proxy: http://127.0.0.1:8999
    name: "Best_Qwen3.6-27B-Q3_K_M"
    description: "-m /models/Qwen3.6-27B-Q3_K_M.gguf --mmproj /models/mmproj-BF16.gguf --ctx-size ${default_ctx} --temp ${temp} --top-p 0.95 --top-k 20 --min-p 0.00  --reasoning-preserve  --no-mmproj-offload --host 0.0.0.0 --fit off -ngl 40  --port 8999"
    cmd: /usr/local/bin/llama-server  -m /models/Qwen3.6-27B-Q3_K_M.gguf --mmproj /models/mmproj-BF16.gguf --ctx-size ${default_ctx} --temp ${temp} --top-p 0.95 --top-k 20 --min-p 0.00  --reasoning-preserve  --no-mmproj-offload --host 0.0.0.0 --fit off -ngl 40  --port 8999
    aliases:
      - "best-Qwen3.6-27B-Q3_K_M"
      - "llama-swap"
    macros:
      "default_ctx": 140000
      "temp": 1.0
    metadata:
      temperature: ${temp}
      note: "The ${MODEL_ID} is running on port 8999 temp=${temp}, context=${default_ctx}"
    env:
      - "NVM_BIN=/root/.nvm/versions/node/v24.17.0/bin"
      - "NVM_DIR=/root/.nvm"
      - "NVM_INC=/root/.nvm/versions/node/v24.17.0/include/node"
      - "PATH=/root/.nvm/versions/node/v24.17.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
    ttl: 60
    unloadTimeout: 30
  "Qwen3.6-27B-Q3_K_M.gguf":
    proxy: http://127.0.0.1:8999
    cmd: /usr/local/bin/llama-server  -m /models/Qwen3.6-27B-Q3_K_M.gguf --ctx-size ${default_ctx}  --temp 1.0  --top-p 0.95 --top-k 20 --min-p 0.00  --reasoning-preserve --host 0.0.0.0 --fit off -ngl 40  --port 8999
    macros:
      "default_ctx": 100000
      "temp": 1.0
    metadata:
      temperature: ${temp}
      note: "The ${MODEL_ID} is running on port 8999 temp=${temp}, context=${default_ctx}"
    env:
      - "NVM_BIN=/root/.nvm/versions/node/v24.17.0/bin"
      - "NVM_DIR=/root/.nvm"
      - "NVM_INC=/root/.nvm/versions/node/v24.17.0/include/node"
      - "PATH=/root/.nvm/versions/node/v24.17.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
    ttl: 60
    unloadTimeout: 10
  "Qwen3.6-27B-Q3_K_M.gguf_mmproj-BF16.gguf_CUDA0":
    proxy: http://127.0.0.1:8999
    cmd: /usr/local/bin/llama-server   -m /models/Qwen3.6-27B-Q3_K_M.gguf --mmproj /models/mmproj-BF16.gguf  --ctx-size ${default_ctx}  --temp ${temp}  --device CUDA0 --top-p 0.95  --top-k 20 --min-p 0.00 --host 0.0.0.0 --port 8999
    macros:
      "default_ctx": 100000
      "temp": 1.0
    metadata:
      temperature: ${temp}
      note: "The ${MODEL_ID} is running on port 8999 temp=${temp}, context=${default_ctx}"
    env:
      - "NVM_BIN=/root/.nvm/versions/node/v24.17.0/bin"
      - "NVM_DIR=/root/.nvm"
      - "NVM_INC=/root/.nvm/versions/node/v24.17.0/include/node"
      - "PATH=/root/.nvm/versions/node/v24.17.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
    ttl: 60
    unloadTimeout: 10
  "Qwen3.6-27B-Q3_K_M.gguf_mmproj-BF16.gguf_CUDA":
    proxy: http://127.0.0.1:8999
    cmd: /usr/local/bin/llama-server   -m /models/Qwen3.6-27B-Q3_K_M.gguf --mmproj /models/mmproj-BF16.gguf  --ctx-size ${default_ctx}  --temp ${temp}  --device CUDA1 --top-p 0.95  --top-k 20 --min-p 0.00 --host 0.0.0.0 --port 8999
    macros:
      "default_ctx": 100000
      "temp": 1.0
    metadata:
      temperature: ${temp}
      note: "The ${MODEL_ID} is running on port 8999 temp=${temp}, context=${default_ctx}"
    env:
      - "NVM_BIN=/root/.nvm/versions/node/v24.17.0/bin"
      - "NVM_DIR=/root/.nvm"
      - "NVM_INC=/root/.nvm/versions/node/v24.17.0/include/node"
      - "PATH=/root/.nvm/versions/node/v24.17.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
  "Qwen3.6-27B-Q3_K_M.gguf_mmproj-BF16.gguf_init":
    proxy: http://127.0.0.1:8999
    cmd: /usr/local/bin/llama-server  -m /models/Qwen3.6-27B-Q3_K_M.gguf --mmproj /models/mmproj-BF16.gguf --ctx-size ${default_ctx} --n-cpu-moe 35 --temp ${temp}  --top-p 0.95 --top-k 20 --min-p 0.00 --port 8999
    macros:
      "default_ctx": 100000
      "temp": 1.0
    metadata:
      temperature: ${temp}
      note: "The ${MODEL_ID} is running on port 8999 temp=${temp}, context=${default_ctx}"
    env:
      - "NVM_BIN=/root/.nvm/versions/node/v24.17.0/bin"
      - "NVM_DIR=/root/.nvm"
      - "NVM_INC=/root/.nvm/versions/node/v24.17.0/include/node"
      - "PATH=/root/.nvm/versions/node/v24.17.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
  "gemma-3-1b-it-q4_k_m.gguf":
    proxy: http://127.0.0.1:8999
    cmd: /usr/local/bin/llama-server  -m /models/gemma-3-1b-it-q4_k_m.gguf --ctx-size ${default_ctx} --n-cpu-moe 35 --temp ${temp}  --top-p 0.95 --top-k 20 --min-p 0.00 --port 8999
    macros:
      "default_ctx": 100000
      "temp": 1.0
    metadata:
      temperature: ${temp}
      note: "The ${MODEL_ID} is running on port 8999 temp=${temp}, context=${default_ctx}"
    env:
      - "NVM_BIN=/root/.nvm/versions/node/v24.17.0/bin"
      - "NVM_DIR=/root/.nvm"
      - "NVM_INC=/root/.nvm/versions/node/v24.17.0/include/node"
      - "PATH=/root/.nvm/versions/node/v24.17.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
    ttl: 60
    unloadTimeout: 30

Cela fonctionne :

J’avais oublier de modifier le script /usr/local/bin/aorus-bridge .

VENDOR_ID="${VENDOR_ID:-0x10de}"   # NVIDIA
# DEVICE_ID: exact PCI device id to match (e.g. 0x2b85 for RTX 5090). Leave
# empty to match ANY NVIDIA display controller by PCI class — this covers the
# whole RTX 50-series (5090=0x2b85, 5060 Ti=0x2d04, …) without a hardcoded list,
# in the same spirit as is_tb_tunneled below. Set DEVICE_ID to pin one board.
DEVICE_ID="${DEVICE_ID:-0x2d04}"
# PCI base class 0x03 = display controller (VGA 0x0300 / 3D 0x0302); selects the
# GPU function and excludes its HDMI-audio function (class 0x0403).
GPU_CLASS_PREFIX="${GPU_CLASS_PREFIX:-0x03}"

Voici pourquoi il ne faut pas de TuxedoOS … perte d’une journée à cause de la suppression du drivers ( plus récent )

# nvidia-smi 

Mon Jul 27 15:15:18 2026       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 610.43.03              KMD Version: 610.43.03     CUDA UMD Version: 13.3     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 3060 ...    Off |   00000000:01:00.0 Off |                  N/A |
| N/A   50C    P0             26W /  115W |       0MiB /   6144MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
|   1  NVIDIA GeForce RTX 5060 Ti     Off |   00000000:05:00.0 Off |                  N/A |
| 45%   36C    P0             15W /  180W |       0MiB /  16311MiB |      1%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+

Pas de crash pour l’instant, la solution est assez stable.