feat(gallery): add Qwen 3D Coder preview

Add the pinned Q8 preview for llama.cpp and document installation.
Keep the HTDemucs source checksum aligned with the fix in #12919.

Assisted-by: Codex:gpt-6

(cherry picked from commit 66b4228310, mudler/LocalAI#12922)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

gallery/index.yaml

@@ -3,6 +3,42 @@
 # written by scripts/models/cmd/modelcatalog and stays only so releases before
 # the cutover keep working. CI fails when it differs from the generated copy:
 #   go run ./scripts/models/cmd/modelcatalog --library gallery/models/index.yaml --out gallery/index.yaml
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: qwen3.5-4b-3d-coder-q8
+  description: |
+    CCORE's early preview fine-tune of Qwen3.5-4B for coding and Three.js
+    scene generation. This text-only Q8_0 build pins a preview
+    checkpoint and uses the embedded chat template. MTP is not enabled.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/Qwen/Qwen3.5-4B
+    - https://huggingface.co/ccore/Qwen3.5-4B-3D-Coder
+  tags:
+    - llm
+    - gguf
+    - coding
+    - preview
+    - cpu
+    - gpu
+  files:
+    - filename: Qwen3.5-4B-3D-Coder-Q8_0.gguf
+      sha256: 666ebe87dcbe4138888f07c4eb1bcb9be82e7a057a5c154c0104f4173690038c
+      uri: https://huggingface.co/ccore/Qwen3.5-4B-3D-Coder/resolve/c8b9b5e565967b81647b9e4c77feaa11782466cc/Qwen3.5-4B-3D-Coder-Q8_0.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 32768
+    known_usecases:
+      - chat
+    options:
+      - use_jinja:true
+    parameters:
+      model: Qwen3.5-4B-3D-Coder-Q8_0.gguf
+      repeat_penalty: 1.0
+      temperature: 0.7
+      top_k: 20
+      top_p: 0.95
+    template:
+      use_tokenizer_template: true
 - url: github:mudler/LocalAI/gallery/virtual.yaml@master
   name: sage-a1.5-q4
   description: |

gallery/models/index.yaml

@@ -4,6 +4,50 @@
 # changes and never edit it by hand. CI fails when it is stale. A build from
 # master reads it as the default model source; a tagged build reads the gallery
 # until the release cutover.
+- kind: model
+  id: qwen3.5-4b-3d-coder-q8
+  name: qwen3.5-4b-3d-coder-q8
+  version: 0.0.0-gallery.c8b9b5e56596
+  description: |
+    CCORE's early preview fine-tune of Qwen3.5-4B for coding and Three.js
+    scene generation. This text-only Q8_0 build pins a preview
+    checkpoint and uses the embedded chat template. MTP is not enabled.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/Qwen/Qwen3.5-4B
+    - https://huggingface.co/ccore/Qwen3.5-4B-3D-Coder
+  tags:
+    - llm
+    - gguf
+    - coding
+    - preview
+    - cpu
+    - gpu
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: Qwen3.5-4B-3D-Coder-Q8_0.gguf
+        sha256: 666ebe87dcbe4138888f07c4eb1bcb9be82e7a057a5c154c0104f4173690038c
+        uri: https://huggingface.co/ccore/Qwen3.5-4B-3D-Coder/resolve/c8b9b5e565967b81647b9e4c77feaa11782466cc/Qwen3.5-4B-3D-Coder-Q8_0.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 32768
+      known_usecases:
+        - chat
+      options:
+        - use_jinja:true
+      parameters:
+        model: Qwen3.5-4B-3D-Coder-Q8_0.gguf
+        temperature: 0.7
+        top_k: 20
+        top_p: 0.95
+        repeat_penalty: 1.0
+      template:
+        use_tokenizer_template: true
 - kind: model
   id: sage-a1.5-q4
   name: sage-a1.5-q4