alexpires.me/a13labs.infra
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

Add multiple Dockerfiles for various Gitea runner profiles and update host vars

Browse Source
This commit is contained in:
Alexandre Pires
2026-07-03 21:52:13 +02:00
parent c6f3b0f829
commit 9a9b36cac6
10 changed files with 27 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ansible/host_vars/gpu-01.lab.alexpires.me.yml
+20 -68
View File
@@ -119,14 +119,6 @@ llama_server_devices:
llama_server_models:
# TODO: replace revision values with pinned commit SHAs for strict reproducibility.
- name: "gpt-oss-20b-q4-0"
model_id: "unsloth/gpt-oss-20b-GGUF"
quant: "Q4_0"
revision: "main"
- name: "gpt-oss-20b-q8-0"
model_id: "unsloth/gpt-oss-20b-GGUF"
quant: "Q8_0"
revision: "main"
- name: "gemma-4-12b-q8-0"
model_id: "unsloth/gemma-4-12b-it-GGUF"
quant: "Q8_0"
@@ -145,27 +137,17 @@ llama_server_models:
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-5-9b-iq4-xs"
model_id: "unsloth/Qwen3.5-9B-GGUF"
quant: "IQ4_XS"
- name: "gpt-oss-20b-q4-0"
model_id: "unsloth/gpt-oss-20b-GGUF"
quant: "Q4_0"
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-5-9b-q8-0"
model_id: "unsloth/Qwen3.5-9B-GGUF"
- name: "gpt-oss-20b-q8-0"
model_id: "unsloth/gpt-oss-20b-GGUF"
quant: "Q8_0"
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-5-27b-mtp-iq4-xs"
model_id: "unsloth/Qwen3.5-27B-MTP-GGUF"
quant: "IQ4_XS"
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-5-35b-a3b-ud-iq4-xs"
model_id: "unsloth/Qwen3.5-35B-A3B-GGUF"
quant: "UD-IQ4_XS"
- name: "qwen3-5-27b-q4-k-m"
model_id: "unsloth/Qwen3.5-27B-GGUF"
quant: "Q4_K_M"
revision: "main"
preset:
ctx-size: 98304
@@ -175,9 +157,15 @@ llama_server_models:
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-6-27b-iq4-xs"
- name: "qwen3-5-9b-q8-0"
model_id: "unsloth/Qwen3.5-9B-GGUF"
quant: "Q8_0"
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-6-27b-q4-k-m"
model_id: "unsloth/Qwen3.6-27B-GGUF"
quant: "IQ4_XS"
quant: "Q4_K_M"
revision: "main"
preset:
ctx-size: 98304
@@ -187,24 +175,6 @@ llama_server_models:
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-6-27b-mtp-ud-q4-0-k-xl"
model_id: "unsloth/Qwen3.6-27B-MTP-GGUF"
quant: "UD-Q4_K_XL"
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-6-27b-mtp-ud-q4-xl"
model_id: "unsloth/Qwen3.6-27B-MTP-GGUF"
quant: "UD-Q4_K_XL"
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-6-35b-a3b-mtp-ud-iq4-xs"
model_id: "unsloth/Qwen3.6-35B-A3B-MTP-GGUF"
quant: "UD-IQ4_XS"
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-6-35b-a3b-mtp-ud-q4-m"
model_id: "unsloth/Qwen3.6-35B-A3B-MTP-GGUF"
quant: "UD-Q4_K_M"
@@ -217,28 +187,6 @@ llama_server_models:
revision: "main"
preset:
ctx-size: 98304
- name: "qwen3-6-35b-a3b-claude47-distilled-iq4-xs"
model_id: "lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-IQ4_XS-GGUF"
quant: "IQ4_XS"
revision: "main"
preset:
ctx-size: 98304
- name: "ministral-3-14b-instruct-iq4-xs"
model_id: "unsloth/Ministral-3-14B-Instruct-2512-GGUF"
quant: "IQ4_XS"
revision: "main"
preset:
ctx-size: 98304
- name: "granite-4-1-30b-iq4-xs"
model_id: "unsloth/granite-4.1-30b-GGUF"
quant: "IQ4_XS"
revision: "main"
preset:
ctx-size: 98304
- name: "devstral-small-2-24b-instruct-iq4-xs"
model_id: "unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF"
quant: "IQ4_XS"
revision: "main"
llama_server_argv_extra:
[
@@ -269,6 +217,10 @@ llama_server_argv_extra:
"-mg",
1,
"--mmap",
"--spec-type",
"draft-mtp",
"--spec-draft-n-max",
2,
]
auto_suspend_enabled: true