gitea/k3s-manifests
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

new qwen

Browse Source
This commit is contained in:
Alexander Rogov
2026-08-25 15:39:31 +03:00
parent b239fc9340
commit 7053071cb2
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
manifests/llama/main.yaml
View File

@@ -27,9 +27,9 @@ spec:
- name: HF_HOME
value: /models/.hf
- name: MODEL_REPO
value: "byteshape/Qwen3.6-35B-A3B-GGUF"
value: "byteshape/Qwen3.8-27B-GGUF"
- name: MODEL_FILE
value: "Qwen3.6-35B-A3B-IQ3_S-3.00bpw.gguf"
value: "Qwen3.8-27B-IQ3_S-3.44bpw.gguf"
# optional, only if you need gated/private models
# - name: HUGGING_FACE_HUB_TOKEN
# valueFrom:
@@ -82,7 +82,7 @@ spec:
image: ghcr.io/ggml-org/llama.cpp:server-vulkan
args:
- "--model"
- "/models/Qwen3.6-35B-A3B-IQ3_S-3.00bpw.gguf"
- "/models/Qwen3.8-27B-IQ3_S-3.44bpw.gguf"
- "--host"
- "0.0.0.0"
- "--port"